江汉大学学报(自然科学版) ›› 2012, Vol. 40 ›› Issue (3): 38-40.

• 数学 • 上一篇    下一篇

2012非线性与线性典型相关分析的对比实验

方建斌1,3,管琼2,王雨春3   

  1. 1.江汉大学数学与计算机科学学院,湖北武汉430056;
    2.湖南理工学院信息与通信工程学院,湖南岳阳414006;
    3.武汉理工大学理学院,湖北武汉430070
  • 收稿日期:2012-04-20 出版日期:2012-06-20 发布日期:2013-11-07
  • 作者简介:方建斌(1974—),男,高级实验师,硕士,研究方向:图像与信息处理
  • 基金资助:
    武汉市科技局科技攻关计划项目(200860423202

ExactNumericThe Contrast Experiment of Canonical Correlation Analysis Between Nonlinear and Linear

FANG Jian-bin1,3,GUAN Qiong2,WANG Yu-chun3   

  1. 1. School of Mathematics and Computer Sciences,Jianghan University,Wuhan 430056,Hubei,China;
    2. College of Information and Communication Engineering,Hunan Institute of Science and Technology,Yueyang 414006,Hunan,China;
    3.School of Science,Wuhan University of Technology,Wuhan 430070,Hubei,China
  • Received:2012-04-20 Online:2012-06-20 Published:2013-11-07

摘要: 线性典型相关分析揭示了两组变量间潜在的线性关系,但实际应用中,变量之间往往还潜在着非线性关系。主要研究非线性典型相关分析算法,揭示变量间潜在的非线性关系,并通过非线性与线性典型相关分析对比实验,验证其优良性能。

关键词: 典型相关分析, 非线性, 核函数

Abstract: Linear canonical correlation analysis reveals potential linear relationship between two groups of variables, but in practical application, there is also potential nonlinear relationship between variables. The paper studies the nonlinear canonical correlation analysis algorithm, reveals variables′ potential nonlinear relationships, and verifies their good performance through the contrast experiment.

Key words: canonical correlation analysis, nonlinear, kemel function

中图分类号: