Share Email Print
cover

Proceedings Paper

Selection of significant samples to reduce the complexity of least-squares support vector machine
Author(s): Giuseppe Di Salvo; Stanisław Jankowski; Ewa Piątkowska-Janko; Paolo Arena
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The least-squares support vector machines (LS-SVM) can be obtained by solving a simpler optimization problem than that in standard support vector machines (SVM). Its shortcoming is the loss of sparseness and this usually results in slow testing speed. Several pruning methods have been proposed to improve the sparseness of a LS-SVM trained on the whole training dataset. A selection of significative samples is proposed to train a LS-SVM on a reduced dataset. A dataset about electrocardiogram (ECG) of 376 patients has been used to assess the proposed algorithm.

Paper Details

Date Published: 28 December 2007
PDF: 10 pages
Proc. SPIE 6937, Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2007, 69371X (28 December 2007); doi: 10.1117/12.784710
Show Author Affiliations
Giuseppe Di Salvo, Univ. degli Studi di Catania (Italy)
Stanisław Jankowski, Warsaw Univ. of Technology (Poland)
Ewa Piątkowska-Janko, Warsaw Univ. of Technology (Poland)
Paolo Arena, Univ. degli Studi di Catania (Italy)


Published in SPIE Proceedings Vol. 6937:
Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2007

© SPIE. Terms of Use
Back to Top