一种基于Morlet小波核的约简支持向量机  

Novel Reduced Support Vector Machine on Morlet Wavelet Kernel Function

作　　者：武方方[1] 赵银亮[1] 

机构地区：[1]西安交通大学新型计算机研究所,西安710049

出　　处：《控制与决策》2006年第8期848-852,856,共6页Control and Decision

基　　金：国家自然科学基金项目(60173066)

摘　　要：针对支持向量机(SVM)的训练数据量仅局限于较小样本集的问题,结合M orlet小波核函数,提出了一种基于M orlet小波核的约简支持向量机(MW RSVM-DC).算法的核心是通过密度聚类寻找聚类中每个簇的边缘点作为约简集合,并利用该约简集合寻找支持向量.实验表明,利用小波核,该算法不仅提高了分类的准确率,而且提高了整体分类效率.To deal with the problem that support vector machine （SVM） is restricted to work well on the small sample sets, based on the Morlet wavelet kernel function, a novel reduced support vector machine on Morlet wavelet kernel function （MWRSVM-DC） is proposed. The presented algorithm focuses on dealing with a sample set through density clustering prior to classifying the samples. After clustering the positive samples and negative samples, the algorithm picks out such samples that locate on the edge of clusters as reduced samples. These reduced samples are treated as the new training sample set used in SVM＇s classifier system. Experiment results show that both the precision and the efficiency of SVM＇s are improved by MWRSVM-DC.

关 键 词：Morlet小波核函数 支持向量机 约倚支持向量机 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]