摘要
针对现有基于析取范式的置信规则库构建方法存在过拟合、未合理利用已知数据信息等问题,提出利用聚类分析构建基于析取范式的置信规则库的新方法.该方法通过对样本数据的输出结果进行聚类分析,获取输出结果的分布特征,以确定系统的规则数、结果评价等级等相关参数,并充分利用样本已知信息,完成基于析取范式的置信规则库构建.为验证所提方法的有效性,选取输油管道和桥梁风险评估两个实验进行验证,结果表明,所提方法构建规则库系统能快速确定系统规模,并获得较高的推理性能.
There are some problems in constructing the disjunctive belief rule base,such as over-fitting,subjective and using the data unreasonable.Therefore,this paper proposes a new method of constructing the disjunctive belief rule base by cluster.This method is used to cluster the output data to obtain the distribution characteristics of the results.In this way,we can make full use of the information of the sample to determine the rule number of the system,believed to be the consequent and other related parameters.Then,we can complete the construction of the disjunctive belief rule base.In order to verify the method is correct and effective,we choose the classic experiments of belief rule base.The one is pipeline leak detection,the other is the bridge risk assessment.The results show that the method proposed in this paper can quickly determine the system scale and have higher reasoning ability.
作者
张婕
傅仰耿
巩晓婷
ZHANG Jie;FU Yanggeng;GONG Xiaoting(College of Mathematics and Computer Science,Fuzhou University,Fuzhou,Fujian 350108,China;Decision Sciences Institute,Fuzhou University,Fuzhou,Fujian 350108,China)
出处
《福州大学学报(自然科学版)》
CAS
北大核心
2019年第4期435-440,共6页
Journal of Fuzhou University(Natural Science Edition)
基金
国家自然科学基金资助项目(71501047
61773123)
关键词
置信规则库
析取范式
规则库构建
证据推理
聚类
K均值
belief rule base
disjunctive paradigm
construct rule base
evidence reasoning
cluster
K-means