基于AFS理论及遗传算法的模糊分类器的设计
首发时间:2008-10-07
摘要:AFS理论是一种新的处理模糊数据的方法,应用AFS理论求得的隶属函数与以往所用的方法相比,具有更精确,最大限度的保留了原始数据信息的特点。本文给出了一种将AFS理论与遗传算法相结合的模糊分类器设计方法:采用AFS理论求出隶属函数,并通过此隶属函数得到分类器的模糊规则,最后将得到的规则应用到遗传算法中进行规则的删减。实验结果说明将两者结合设计出的模糊分类器不仅规则少,而且分类率高。
For information in English, please click here
Fuzzy classifier Based on AFS theory and Genetic Algorithms
Abstract:AFS theory is a new research method for dealing with fuzzy data, compared to the other methods, the membership calculated by AFS theory is more accurate and characteristic of preserving original information to a maximum extent. In this paper, a new classifier has been proposed based on AFS theory and Genetic Algorithms: Membership function is calculated by AFS theory. Then, with the membership function, we get rules that are used to classify the sample. Finally, these rules are selected by Genetic Algorithms. The simulate result shows that the proposed fuzzy classifier based on AFS theory and Genetic Algorithms has few rules and high classification rate.
Keywords: AFS fuzzy logic Genetic Algorithms Fuzzy classifier
基金:
论文图表:
引用
No.2457530802912233****
同行评议
共计0人参与
勘误表
基于AFS理论及遗传算法的模糊分类器的设计
评论
全部评论0/1000