一种改进的模糊支持向量机隶属度确定方法  

An improved method for determining the membership of fuzzy SVM

作　　者：刘成忠[1,2] 韩虎[3] 黄高宝 

机构地区：[1]甘肃农业大学信息科学技术学院,兰州730070 [2]甘肃省干旱生境作物学重点实验室,兰州730070 [3]兰州交通大学数理与软件工程学院,兰州730070

出　　处：《四川大学学报（自然科学版）》2011年第3期546-550,共5页Journal of Sichuan University(Natural Science Edition)

基　　金：甘肃省自然科学基金(096RJZA004);甘肃省教育厅科研基金(0902-04);甘肃省科技支撑计划(1011NKCA058)

摘　　要：为了克服支持向量机方法对于噪声或异常样本敏感的问题,本文研究基于粗糙集理论的粗糙单类支持向量机,提出一种改进的模糊支持向量机隶属度确定方法.该算法首先利用粗糙集思想构造一个最小粗糙球,分别得到对应粗糙球的上近似、下近似与边界区域,然后依据样本在超球中的位置对分布在下近似、边界域和粗糙球以外的样本,分别采用三种不同的方式计算其各自的隶属度.最后通过对比实验表明,与传统支持向量机方法相比,本文提出的改进方法使模糊支持向量机具有更好的抗噪性能及分类能力.In order to overcome the problem that support vector machine is sensitive to noises and outli- ers, rough one-class support vector machine based on rough set is researched and an improved method for determining the membership of fuzzy support vector machine is proposed in the paper. At first, the smallest rough sphere which encloses the data points is constructed. And then, those data points are divided into upper approximation region, low approximation region and boundary region by the distance between data points and the center of the rough sphere. At last, several comparative experiments using synthetic and real life data set show the performance and the validity of the method.

关 键 词：模糊支持向量机 隶属度 粗糙集 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]