Fisher和支持向量的综合分类器  

Fisher-support vector classifier

作　　者：安文娟[1] 孙德山[1] 

机构地区：[1]辽宁师范大学数学学院,辽宁大连116029

出　　处：《计算机工程与应用》2010年第29期183-185,共3页Computer Engineering and Applications

基　　金：辽宁省高等学校科研项目资助(No.2008343)

摘　　要：结合Fisher判别分析和支持向量机的优点,提出了一种新的分类算法—Fisher-SV分类器(简称FSVC)。该分类器的核心思想就是寻找最优分类面的法向量w*,使得样本向量在w*上做投影后,不仅使分类间隔达到最大,而且使类内离散程度尽可能小。对于线性情况,可以转化为传统的支持向量机求解,而不需要设计新的求解算法。对于非线性情况,利用再生核理论得出新的求解算法。实验结果表明,该分类器具有很高的准确度和可靠性。With combination of advantages of both Fisher discriminant analysis and support vector machines,this paper develops an improved classification algorithm,called Fisher-Support Vector Classifier.The central idea is that the vector w^＊ of the optimal hyperplane is found along which the samples are projected such that the margin is maximized while within-class scatter is kept as small as possible.In linear case,it can be converted to traditional Support Vector Machines（SVM） to solve and doesn＇t need to design new algorithms.In nonlinear case, a new algorithm is produced by the reproducing kernel theory. The test result shows that the Fisher-Support Vector Classifier established has a high accuracy and reliability.

关 键 词：FISHER判别 支持向量机 分类间隔 类内离散程度 再生核理论 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]