University of Sussex
Browse

File(s) not publicly available

A geometric method to improve the performance of the support vector machine

journal contribution
posted on 2023-06-08, 09:20 authored by Peter Williams, Sheng Li, Jianfeng Feng, Si Wu
The performance of a support vector machine (SVM) largely depends on the kernel function used. This letter investigates a geometrical method to optimize the kernel function. The method is a modification of the one proposed by S. Amari and S. Wu. Its concern is the use of the prior knowledge obtained in a primary step training to conformally rescale the kernel function, so that the separation between the two classes of data is enlarged. The result is that the new algorithm works efficiently and overcomes the susceptibility of the original method

History

Publication status

  • Published

Journal

IEEE Transactions on Neural Networks

ISSN

1045-9227

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Issue

3

Volume

18

Page range

942-947

Department affiliated with

  • Informatics Publications

Notes

Originality: Significantly improved a well-known method in machine learning for optimizing the kernel function of support vector machines, the current state-of-art pattern recognition method. Rigor: Applied the Information Geometry method to analyze the behaviours of kernel mapping. The new method was tested by real-world problem. Significance: this new method significantly improves the performance of the previous one, and makes this type of geometry-based method more robust and easier to use. Outlet/Citation: Top Engineering/Machine Learning journal. In press.

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2013-02-20

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC