CQUniversity
Browse

Automatic parameter selection for polynomial kernel

Download (420.7 kB)
conference contribution
posted on 2017-12-06, 00:00 authored by A B M Shawkat Ali, K Smith
Kernel is the heart of kernel based learning. To choose an appropriate parameter for a specific kernel is an important research issue in the data mining area. In this paper we propose an automatic parameter selection approach for polynomial kernel. The algorithm is tested on Support Vector Machines (SVM). The parameter selection is considered on the basis of prior information of the data distribution and Bayesian inference. The new approach is tested on different sizes of benchmark datasets with binary class problems as well as multi class classification problems.

History

Parent Title

Proceedings of the IEEE International Conference on Information Reuse and Integration, Las Vegas, 27-29 Oct., 2003.

Start Page

243

End Page

249

Number of Pages

7

Start Date

2003-01-01

ISBN-10

0780382420

Location

Las Vegas

Publisher

IEEE

Place of Publication

New Jersey

Peer Reviewed

  • Yes

Open Access

  • No

External Author Affiliations

Monash University;

Era Eligible

  • Yes

Name of Conference

IEEE International Conference on Information Reuse and Integration

Usage metrics

    CQUniversity

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC