直觉模糊信息系统的规则提取方法  

RULE EXTRACTION METHOD OF INTUITIONISTIC FUZZY INFORMATION SYSTEM

作　　者：李其轩 路艳丽[2] 

机构地区：[1]空军工程大学航空航天工程学院,陕西西安710038 [2]空军工程大学防空反导学院,陕西西安710051

出　　处：《计算机应用与软件》2015年第1期251-254,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(60773209;61272011)

摘　　要：规则获取是直觉模糊推理系统面临的难题之一。针对直觉模糊环境下的规则提取问题,首先给出直觉模糊信息系统的定义,分析决策属性与条件属性及条件属性对应的语言值之间所蕴含的直觉模糊逻辑关系,以此构造信息系统的逻辑关系。进而,通过融合信息系统中具有相同决策值的数据行所表达的决策规则信息,给出包含置信度和非置信度的直觉模糊规则可信度计算方法,并分两种情况讨论基于直觉模糊可信度的规则筛选方法。在此基础上提出直觉模糊信息系统的规则提取算法,最后用具体算例验证了算法的可用性和有效性。Rules acquisition is one of the difficuh problems the intuitionistic fuzzy inference system encounters. For the problem of rules extraction in intuitionistic fuzzy environment, the definition of the intuitionistic fuzzy system is given at first, and the decision attributes and the condition attributes as well as the intuitionistic fuzzy logic relation implicated among the linguistic values corresponding to condition attributes are analysed as well, and the logic relation of information system is constructed in this account. Furthermore, by integrating the information with decision rules expressed by the data lines having same decision value in information system, the calculation method in raged to the reliability of intuitionistic fuzzy rules including confidence and non-confidence degrees is provided, and the rules screening method based on intuitionistic fuzzy reliability is also discussed in two situations. Based on that, the rule extraction algorithm of intuitionistic fuzzy information system is put forward. Finally, through specific numerical example the usability and validity of the algorithm are verified.

关 键 词：直觉模糊集 信息系统 规则提取 可信度 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]