基于线性规划的ν-支持向量机分类器  

ν-Support vector machine classifier based on linear programming

作　　者：宋杰[1] 唐焕文[1] 

机构地区：[1]大连理工大学应用数学系

出　　处：《大连理工大学学报》2005年第2期303-307,共5页Journal of Dalian University of Technology

基　　金：国家自然科学基金资助项目(重点项目:90103033).

摘　　要：Scho..lkopf等提出的基于二次规划的ν-支持向量机(ν-SVM)与标准SVM相比,其优势在于可以控制支持向量的数目和误差,但由于增加了模型的复杂性,限制了其应用.为此,构造了一种基于线性规划的ν-SVM分类器,模型简单,参数ν具有明确的意义,同样可以控制支持向量的数目和误差,直接利用比较成熟的线性规划算法.数值实验表明,该方法ν-SVM的训练速度要比基于二次规划的ν-SVM快得多,而分类效果两者相当.The ν-support vector machine (ν-SVM) classifier proposed by (Sch(o··)lkopf) has the advantage of controlling numbers of support vectors and errors compared to regular SVM. However, its formulation is more complicated which confines its applications. A new and more simple ν-SVM classifier based on linear programming is presented. The parameter also has implicit sense of controlling numbers of support vectors and errors. Furthermore the authors can use effective linear programming solvers available. Numerical tests show that the ν-SVM based on linear programming is much faster than original ν-SVM and performs as accurately as the original one.

关 键 词：支持向量机分类器 SVM分类器 线性规划算法 二次规划 数值实验 直接利用 训练速度 分类效果 复杂性 误差 控制 模型 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TP391.4[自动化与计算机技术—计算机科学与技术]