一种基于训练数据的迭代改进核函数

周志祥; 韩逢庆

一种基于训练数据的迭代改进核函数

周志祥¹,
韩逢庆²

1.
重庆交通大学土木建筑学院,重庆 400074;2.重庆交通大学理学院,重庆 400074

基金项目: 国家自然科学基金资助项目(50578168);重庆市自然科学基金资助项目(2007BB2396)

详细信息

作者简介:
周志祥(1958- ),男,四川遂宁人,教授,博士,博士生导师(联系人.Tel:+86-23-62652922;E-mail:zhixiangzhou@cqjtu.edu.cn);韩逢庆(1968- ),男,重庆人,教授,博士(E-mail:hanfengqing@cqit.edu.cn).

中图分类号: O241.82
计量
- 文章访问数: 2247
- HTML全文浏览量: 112
- PDF下载量: 683
- 被引次数: 0
出版历程
- 收稿日期: 2008-07-18
- 修回日期: 2008-12-03
- 刊出日期: 2009-01-15

An Iterative Modified Kernel Based on Training Data

ZHOU Zhi-xiang¹,
HAN Feng-qing²

1.
College of Civil Engineering and Architecture, Chongqing Jiaotong University, Chongqing 400074, P. R. China;

摘要

摘要: 为提高支持向量机性能，提出一种支持向量机核函数的迭代改进新算法．利用与数据有关的保角映射，使核函数包含了全部学习样本的信息，即核函数具有数据依赖性．基本核函数的参数可取随机初值，通过对核函数进行多次迭代改进，直至得到满意的学习效果．与传统方法相比，新算法不需要筛选核函数的参数．对一元连续函数和强地震事件的仿真计算结果表明，改进SVR（support vector regression）的学习效果优于传统方法，并且随着迭代次数的增加，学习风险下降收敛，收敛速度依赖于传统方法的基本参数和改进方法的参数．
- 支持向量回归 /
- 数据依赖 /
- 核函数 /
- 迭代
Abstract: In order to improve the performance of a support vector regression, a new method for modified kernel function is proposed. In this method the information of whole samples is included in kernel function by conformal mapping. So the kernel function is data-dependent. With random initial parameter of kernel function, the kernel function is modified repeatedly until a satisfactory effect is achieved. Compared with the conventional model, the improved approach needs not selecting parameters of kernel function. Simulation was finished for one-dimension continue function and strong earthquake event. The results show that the improved approach has better learning ability and forecasting precision than the traditional model. With the iteration number increasing, the figure of merit will decrease and converge. The speed of convergence depends on the parameters in the algorithm.
- support vector regression /
- data-dependent /
- kernel function /
- iteration

HTML全文

参考文献(12)

[1]	Vapnik V. 统计学习理论的本质[M].张学工译. 北京:清华大学出版社, 2000.
[2]	Scholkopf B, Sung K, Burges C. Comparing support vector machines with Gaussian kernels to radial basis function classifiers[J].IEEE Trans Signal Processing,1997,45(11):2758-2765. doi: 10.1109/78.650102
[3]	Perez-Cruz F , Navia-Vazquez A , Figueiras-Vidal A R,et al.Empirical risk minimization for support vector classifiers[J].IEEE Trans on Neural Networks,2003,14(2):296-303. doi: 10.1109/TNN.2003.809399
[4]	Belhumeur P N, Hespanha J P, Kriegman D J.Eigenfaces vs Fisherfaces：Recognition using class specific linear projection[J].IEEE Trans on Pattern Analysis and Machine Intelligence,1997,19(7):711-720. doi: 10.1109/34.598228
[5]	CAO Li-juan, Tay Francis E H. Financial forecasting using support vector machines[J]. Neural Computing & Applications,2001,10(2):184-192.
[6]	Tay Francis E H, CAO Li-juan. ε-Descending support vector machines for financial time series forecasting[J].Neural Processing Letters,2002,15(2) :179-195. doi: 10.1023/A:1015249103876
[7]	YANG Hai-qin, CHAN Lai-wan, King Irwin. Support vector machine regression for volatile stock market prediction[A]. In:Yin H, Allinson N, Freeman R,et al,Eds.Proceedings of Intelligent Data Engineering and Automated Learning[C].Berlin:Springer-Verlag, 2002. 319-396.
[8]	Smola A J, Schlkopf B， MLler K R.The connection between regularization operators and support vector kernels[J].Neural Network,1998,11(4):637-649. doi: 10.1016/S0893-6080(98)00032-X
[9]	Amari S, Wu Si. Improving support vector machine classifiers by modifying kernel functions[J].Neural Networks,1999,12(6):783-789. doi: 10.1016/S0893-6080(99)00032-5
[10]	LIANG Yan-chun, SUN Yan-feng. An improved method of support vector machine and its applications to financial time series forcesting[J].Progresss in Natural Science,2003,13(9):696-700.
[11]	Colin C. Kernel methods:a survey of current techniques[J].Neurocomputing,2002,48:63-84. doi: 10.1016/S0925-2312(01)00643-9
[12]	马润勇, 彭建兵. 震级与破裂尺度及位错量关系的讨论[J].西北大学学报(自然科学版), 2006,36(5):799-802.

施引文献

资源附件(0)

访问统计

计量

文章访问数: 2247
HTML全文浏览量: 112
PDF下载量: 683
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

一种基于训练数据的迭代改进核函数

作者简介:
周志祥(1958- ),男,四川遂宁人,教授,博士,博士生导师(联系人.Tel:+86-23-62652922;E-mail:zhixiangzhou@cqjtu.edu.cn);韩逢庆(1968- ),男,重庆人,教授,博士(E-mail:hanfengqing@cqit.edu.cn).

计量

An Iterative Modified Kernel Based on Training Data

计量

目录

留言板

一种基于训练数据的迭代改进核函数

作者简介: 周志祥(1958- ),男,四川遂宁人,教授,博士,博士生导师(联系人.Tel:+86-23-62652922;E-mail:zhixiangzhou@cqjtu.edu.cn);韩逢庆(1968- ),男,重庆人,教授,博士(E-mail:hanfengqing@cqit.edu.cn).

计量

出版历程

An Iterative Modified Kernel Based on Training Data

计量

出版历程

目录

作者简介:
周志祥(1958- ),男,四川遂宁人,教授,博士,博士生导师(联系人.Tel:+86-23-62652922;E-mail:zhixiangzhou@cqjtu.edu.cn);韩逢庆(1968- ),男,重庆人,教授,博士(E-mail:hanfengqing@cqit.edu.cn).