机构地区: 四川大学计算机学院
出 处: 《信息技术》 2003年第3期3-6,88,共5页
摘 要: 提出使用标准阈值α和适用性阈值β(α∈ [0 ,1]且 β∈ [0 ,1]) ,从模糊分类问题的模糊化数据中产生模糊规则。用这个方法处理“星期六早晨问题” ,比现有方法 ,分类准确率更高 ,而产生模糊规则较少。 In this article, a new method for generating fuzzy rules from numerical data for handling fuzzy classification problems based on the level threshold value α and the application threshold value β is presented, where α∈ and β∈. This method deals with the 'Saturday Morning Problem '. The method has a higher classification accuracy rate and generates fewer fuzzy rules than the existing methods.
领 域: [自动化与计算机技术] [自动化与计算机技术]