J4

• • 上一篇    下一篇

支持向量机的数据依赖型核函数改进算法

孙延风1,2, 梁艳春2   

  1. 1. 吉林大学数学学院, 长春 130012; 2. 吉林大学计算机科学与技术学院, 长春 130012
  • 收稿日期:2003-01-04 修回日期:1900-01-01 出版日期:2003-07-26 发布日期:2003-07-26
  • 通讯作者: 梁艳春

An Improved Method for Kernel Function withData-Dependent Type of Support Vector Machine

SUN Yan-feng1,2, LIANG Yan-chun2   

  1. 1. College of Mathematics Science, Jilin University, Changchun 130012, China; 2. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2003-01-04 Revised:1900-01-01 Online:2003-07-26 Published:2003-07-26
  • Contact: LIANG Yan-chun

摘要: 基于信息几何理论, 提出一种新的支持向量机核函数改进算法. 利用与数据有关的保角映射, 使核函数具有数据依赖性. 对股票价格数据进行预测的数值模拟结果表明, 改进算法比常规模型具有更好的预测精度.

关键词: 支持向量机, 数据依赖, 信息几何, 保角映射

Abstract: A novel improved method for kernel function of support vector machine(SVM) is proposed based on the information geometry theory. The kernel function is modified by means of a conformal mapping, which makes the kernel function data-dependent. The simulated results on the prediction of stock price show that the improved approach possesses better forecasting precision than the conventional models.

Key words: support vector machine, data dependent, information geometry, conformal mapping

中图分类号: 

  • TP183