An information-geometrical approach to kernel construction in SVM and its application in soft-sensor modeling

Author

An, Wen-Sen ; Sun, Yan-Guang

Author_Institution

Dept. of Autom., Univ. of Sci. & Technol. of China, China

Volume

7

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

4356

Abstract

The method of soft-sensor modeling based on support vector machine is first analysed. Geometry of kernel function is studied from information geometry perspective in view of important influence of kernel´s type on the performance of support vector machine. Then the kernel function is constructed in data-dependent way in order to improve the performance of support vector machine. Simulation results for both artificial data and real application show the effectiveness of the proposed method.

Keywords

intelligent sensors; learning (artificial intelligence); neural nets; regression analysis; support vector machines; SVM; information geometry; kernel function geometry; soft-sensor modeling; support vector machine; Artificial neural networks; Construction industry; Design automation; Information geometry; Kernel; Metals industry; Solid modeling; Sun; Support vector machine classification; Support vector machines; Information geometry; kernel function; soft-sensor modeling; support vector machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527704

Filename

1527704