基于主次原型超平面最接近支持向量机  

The Proximal Support Vector Machine Based on the Primary and Secondary Prototypal Hyperplanes

作　　者：马波[1] 王正群[1] 侯艳平[1] 沈洁[1] 

机构地区：[1]扬州大学信息工程学院,江苏扬州225009

出　　处：《计算机工程与科学》2009年第1期148-150,共3页Computer Engineering & Science

基　　金：江苏省高校自然科学基础研究项目(07KJB520133;05KJB520152);国家自然科学基金资助项目(60774017)

摘　　要：基于广义特征值的最接近支持向量机GEPSVM是一种新的具有与SVM性能相当的两类分类方法,通过求解广义特征值来获得两个彼此不平行的拟合两类样本的超平面,其决策规则是将测试样本归为距其最近的超平面所在的类。然而,该规则在某些情形会导致较差的分类结果。对此,本文提出了在利用GEPSVM产生一个主原型超平面的基础上,再利用主原型超平面及它类样本的信息构造一个次原型超平面,形成一个由主次原型超平面共同决策的最接近支持向量机。该方法不仅简单且易于实现,而且具有较GEPSVM更优的分类性能。在UCI数据集上的实验验证了它的有效性。A binary classifier termed as proximal support vector machine via generalized eigenvalues （GEPSVM） is proposed recently. It aims to obtain two nonparallel planes generated from their corresponding generalized eigenvalue problem and has an equivalent classification efficiency to SVM. In nature, GEPSVM attempts to fit two-class points using two planes. For an unknown sample, according to the decision rule of GEPSVM, it will be assigned to the closest planes. In fact, this rule, in many cases, may result in wrong classification. In this paper, based on GEPSVM, a new proximal support vector machine based on the primary and secondary prototypal hyperplanes is propose＆ First, it produces the primary prototypal hyperplanes by GEPSVM, and then, it produces the secondary prototypal hyperplanes using primary prototypal hyperplanes and the information of other classes. This method is not only simple and easy-to-realize, but also improves the classification accuracy of GEPSVM, which has been validated on real UCI datasets.

关 键 词：最接近支持向量机 广义特征值 原型超平面 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]