一种基于高斯分布的SVM核参数选择方法  

A SVM Kernel Parameter Selection Approach Based on the Gauss Distribution

作　　者：孔晓斌[1] 曹棣[2] 张素兰[1] 

机构地区：[1]太原科技大学计算机科学与技术学院,山西太原030024 [2]太原理工大学计算机与软件学院,山西太原030024

出　　处：《中北大学学报（自然科学版）》2008年第4期343-346,共4页Journal of North University of China(Natural Science Edition)

摘　　要：针对支持向量机(Support Vector Machine,SVM)研究中的核选择问题,在Amari和Wu提出的保形变换方法的基础上,提出了一种基于高斯分布的SVM核参数选择方法.分析了确定数据分布特征的重要性,给出了判断数据呈高斯分布的方法,探讨了SVM核函数及其参数选择与数据分布的相关性.在Matlab实验环境下,采用两组数据集进行了数值仿真,仿真结果说明了本文所提方法的可行性与有效性.Aiming at the problem of kernel selection for support vector machine （SVM）, a new method for selecting the kernel parameter was presented. It was based on the Gauss distribution and conformal transformation method presented by Amari and Wu. The importance of determining the characteristics of data distribution was analyzed, and an approach to determine the Gauss distribution was presented. The correlation of data distribution and how to select the kernel parameter was discussed. Two groups of data sets were used to carry on the numerical simulation by Matlab. The simulating results demonstrate the feasibility and the effectiveness of the presented approach.

关 键 词：支持向量机 核参数选择 高斯分布 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]