摘要
介绍了支持向量机算法的基本思想、数据分类的概念,分析了传统支持向量机算法的一般特性。用Libsvm工具箱实现了基于SVM算法的分类器设计,并用公共数据库中的数据集对设计的分类器进行了测试,重点针对训练样本的选择、参数的影响选择与优化问题进行了研究。实验结果表明,在应用支持向量机算法做数据分类时,选择合适的训练样本和参数有利于提高分类器的准确度。
The basic idea of the Support Vector Machine （SVM） algorithm and the concept of data classification is introduced. An analysis of the traditional SVM general characteristics is also made. A classifier based on the SVM algorithm is designed with Libsvm toolbox, and is tested by data set from public databases. The paper focuses on the selection of training samples, as well as the effect and optimization of parameters. The experimental results show that suitable training samples and parameters help to improve the accuracy of the classifier.
出处
《电子科技》
2015年第4期23-26,共4页
Electronic Science and Technology
