Geometric approach of quasi-linear kernel composition for support vector machine

研究成果: Conference contribution

7 引用 (Scopus)

抜粋

This paper proposes a geometric way to construct a quasi-linear kernel by which a quasi-linear support vector machine (SVM) is performed. A quasi-linear SVM is a SVM with quasi-linear kernel, in which the nonlinear separation boundary is approximated by using multi-local linear boundaries with interpolation. However, the local linearity extraction for the composition of quasi-linear kernel is still an open problem. In this paper, according to the geometric theory, a method based on piecewise linear classifier is proposed to extract the local linearity in a more precise and efficient way. We firstly construct a function set including multiple linear functions and each of those functions reflects one part of linearity of the whole nonlinear separation boundary. Then the obtained local linearity is added as prior information into the composition of quasi-linear kernel. Experimental results on synthetic data sets and real world data sets show that our proposed method is effective to improve classification performances.

元の言語English
ホスト出版物のタイトルProceedings of the International Joint Conference on Neural Networks
出版者Institute of Electrical and Electronics Engineers Inc.
2015-September
ISBN(印刷物)9781479919604, 9781479919604, 9781479919604, 9781479919604
DOI
出版物ステータスPublished - 2015 9 28
イベントInternational Joint Conference on Neural Networks, IJCNN 2015 - Killarney, Ireland
継続期間: 2015 7 122015 7 17

Other

OtherInternational Joint Conference on Neural Networks, IJCNN 2015
Ireland
Killarney
期間15/7/1215/7/17

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

フィンガープリント Geometric approach of quasi-linear kernel composition for support vector machine' の研究トピックを掘り下げます。これらはともに一意のフィンガープリントを構成します。

  • これを引用

    Li, W., & Furuzuki, T. (2015). Geometric approach of quasi-linear kernel composition for support vector machine. : Proceedings of the International Joint Conference on Neural Networks (巻 2015-September). [7280384] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2015.7280384