標題: Multiclass support vector classification via coding and regression
作者: Chen, Pei-Chun
Lee, Kuang-Yao
Lee, Tsung-Ju
Lee, Yuh-Jye
Huang, Su-Yun
資訊工程學系
Department of Computer Science
關鍵字: Kernel map;Multiclass classification;Output code;Regularization;Support vector machine;Support vector regression
公開日期: 1-Mar-2010
摘要: The multiclass classification problem is considered and resolved through coding and regression. There are various coding schemes for transforming class labels into response scores. An equivalence notion of coding schemes is developed, and the regression approach is adopted for extracting a low-dimensional discriminant feature subspace. This feature subspace can be a linear subspace of the column span of original input data or kernel-mapped feature data. The classification training and prediction are carried out in this feature subspace using a linear classifier, which lead to a simple and computationally light but yet powerful toolkit for classification. Experimental results, including prediction ability and CPU time comparison with LIBSVM, show that the regression-based approach is a competent alternative for the multiclass problem. (C) 2009 Elsevier B.V. All rights reserved.
URI: http://dx.doi.org/10.1016/j.neucom.2009.11.005
http://hdl.handle.net/11536/5799
ISSN: 0925-2312
DOI: 10.1016/j.neucom.2009.11.005
期刊: NEUROCOMPUTING
Volume: 73
Issue: 7-9
起始頁: 1501
結束頁: 1512
Appears in Collections:Articles


Files in This Item:

  1. 000276481200044.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.