An approach for kernel selection based on data distribution

Authors:
Wenjian Wang;Jinling Guo;Changqian Men
Affiliations:
Institute of System Engineering, Faculty of Computer and Information Technology, Shanxi University, Taiyuan, P.R. China;Department of Information and Engineering, Business College of Shanxi University, Taiyuan, P.R. China;Institute of System Engineering, Faculty of Computer and Information Technology, Shanxi University, Taiyuan, P.R. China
Venue:
RSKT'08 Proceedings of the 3rd international conference on Rough sets and knowledge technology
Year:
2008

Citing 7
Cited 1

The nature of statistical learning theory

The nature of statistical learning theory
Geometry and invariance in kernel based methods

Advances in kernel methods
Support vector machines for dynamic reconstruction of a chaotic system

Advances in kernel methods
Evaluating the Generalization Ability of Support Vector Machines through the Bootstrap

Neural Processing Letters
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
A Probabilistic Framework for SVM Regression and Error Bar Estimation

Machine Learning
Predicting Time Series with Support Vector Machines

ICANN '97 Proceedings of the 7th International Conference on Artificial Neural Networks

Granular support vector machine based on mixed measure

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a data based kernel selection approach, which utilizes the geometry distribution of data. Once the approximate distribution can be confirmed as a special one like circle, cirque, sphere cylinder, et al, some known kernel functions corresponding to the special distribution can then be used. Four datasets are used to verify the presented approach, and simulation results demonstrate the rationality and effectiveness of the presented approach.