摘要
针对两类分类问题中样本点数量多、类别模糊且有孤立野点的情况,论文在类中心向量方法的基础上,提出了一种基于类中心思想的去边缘模糊支持向量机,该方法用类中心思想预先去掉那些可能不是支持向量的点,并采用降半哥西分布作为隶属度,使其适合模糊分类的性能特点。从理论和实证分析两个方面将该方法与线性可分SVM及模糊SVM进行了对比分析,结果显示该方法不但大大减少了训练点数目,从而减小了内存和计算量,提高了训练速度,而且减少了孤立野点对支持向量分类机的影响。
A new Fuzzy Support Vector Machine of Dismissing Margin(DMFSVM) based on the idea of class-center is proposed aiming at the outliers and noises appearing in the large quantity samples with fuzzy membership.The new algorithm has weeded out some training samples which isn't possible support vectors and adopts a fuzzy membership function of decreasing semi-Cauchy type.Experimental results show that the number of training samples is reduced, which means that the consumption of EMS memory is decreased and the amount of computation is reduced,but training speed is increased.And it can also avoid the bad affect of the outliers and noises in the training samples.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第22期146-149,共4页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(编号:60373090)
航天基金(编号:021.3jw0504)
关键词
推广的最大间隔法
模糊支持向量机
模糊因子
类中心
去边缘方法
generalized method of maximal margin,fuzzy support vector machine,fuzzy membership function,the method of class-center,the method of dismissing margin