基于云模型定性规则推理的分类方法  

Classification Method Based on Qualitative Rule Reasoning of Cloud Model

作　　者：杨丽彬[1] 李海林[2] 

机构地区：[1]龙岩学院数学与计算机科学学院,福建龙岩364012 [2]大连理工大学系统工程研究所

出　　处：《计算机时代》2011年第3期4-6,共3页Computer Era

摘　　要：根据粗糙集原理和模糊集理论,提出了一种基于云模型定性规则推理的分类方法,利用云的相关理论获得多条件单规则中包含隶属度的决策表,结合模糊模式识别技术进行样本分类。针对一些数据对象分别隶属于不同类别的情况,用定性概念来代替模糊集中的定量数据并建立二元关系,能对连续型数据进行更为简单合理的"软"分类,从而使基于定性概念的算法模型符合人类思维方式。According to the theories of rough set and fuzzy set,we propose a classification method based on the qualitative rule reasoning of cloud model.It uses cloud correlation theory to obtain a decision table consisting of multiple conditions and single rule with membership degree,combines the technology of fuzzy pattern recognition to classify samples.For some data objects attaching to different classes,we use the qualitative concept to replace the quantitative data in fuzzy set and build the binary relation,which can conduct more simple and reasonable soft classification on continuous data,so as to make the algorithm model based on qualitative concept conform to human thinking mode.

关 键 词：云模型 规则推理 分类算法 定性概念 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]