基于半模糊核聚类的模糊超球支持向量机多类分类方法  

Fuzzy Sphere Support Vector Machine Multi-class Classification Based on Semi-fuzzy Kernel Clustering

作　　者：郭雪松[1] 孙林岩[1] 刘哲[1] 

机构地区：[1]西安交通大学管理学院

出　　处：《系统管理学报》2007年第6期664-668,共5页Journal of Systems & Management

基　　金：863项目(2003AA413033);国家自然科学基金重点资助项目(70433003)

摘　　要：针对支持向量机(SVM)在大类别模式分类中存在的问题,提出了一种基于半模糊核聚类的超球SVM分类方法。该方法基于半模糊核聚类生成模糊类,利用模糊类提供的边缘样本信息,利用超球SVM进行多类分类,从而有效提高分类器的性能。实验表明,该方法比传统方法具有更高的速度和精度。Aimed at the problems of support vector maehines（SVM） for multi-class pattern recognition with large number of catalogs, a new method of hyper-sphere SVM multi-class classification based on semifuzzy kernel clustering is proposed. The new method defines confusion classes based on semi-fuzzy kernel elustering and sphere support vector machines for the multi-class classification using the information of boundary of confusion classes so as to improve the performance of SVM efficiently. Experimental results indicate that the new method yields higher precision and speed than classical SVM multi-class classification methods.

关 键 词：支持向量机 多类分类 半模糊核聚类 超球 

分 类 号：N945.25[自然科学总论—系统科学] O231.5[理学—运筹学与控制论]