摘要
由于使用单一且固定的核函数,传统的核分类算法不能有效地适应复杂的数据集合,导致分类性能下降.本文提出一种基于双核复合的分类算法ODKC(Optimal Double-Kernel Combination)的构造框架,通过融合两个基本核函数的映射来构造目标核函数.研究了双核复合的三种典型方式,并把这三种复合方式纳入到统一的框架下处理.论文以核与数据的匹配性度量KTA(KernelTarget Alignment)以及分类性能验证了所提算法的有效性.
Traditional kernelised classification methods could not perform well sometimes because of using a single and fixed kernel, especially on some complicated data sets. In this paper, a novel optimal double-kernel combination （ODKC） method is pro- posed for complicated classification tasks. Firstly, data are mapped by two basic kernels into different feature spaces respectively, and then three kinds of optimal composite kernels are constructed by integrating information of the two feature spaces. Comparative ex- periments demonstrate the effectiveness of our methods.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2012年第2期260-265,共6页
Acta Electronica Sinica
基金
国家自然科学基金(No.60775011
No.61005001)
关键词
核方法
双核复合
分类
kernel methods
double-kernel combination
classification
