一种基于环形分布的SVM核选择方法

A Kernel Selection Approach Based on the Characteristics of Cirque Distribution

核选择是支撑向量机(Support Vector Machine,SVM)研究中的核心问题之一.文章提出了一种基于数据分布特征的SVM核函数选择方法.分析了几种常用核函数的性能,提出了判断数据呈环形分布的方法,探讨了SVM核函数及其参数选择与数据分布的相关性.数值实验说明了该方法的可行性与有效性.

The kernel selection is one of the key problems for support vector machine（SVM）.A new way to select the kernel function is presented.It is based on the characteristics of data distribution.We analyse the existing kernel function and present an approach to determining cirque distribution.And then based on determining cirque distribution,it is discussed how to select the kernel function.The simulation experiments demonstrate the feasibility and the effectiveness of the presented approach.