Carregant...
Carregant...

Vés al contingut (premeu Retorn)

A study on output normalization in multiclass SVMs

Autor
Gonzalez-Abril, L.; Velasco, F.; Angulo, C.; Ortega, J.
Tipus d'activitat
Article en revista
Revista
Pattern recognition letters
Data de publicació
2013-02-01
Volum
34
Número
3
Pàgina inicial
344
Pàgina final
348
DOI
https://doi.org/10.1016/j.patrec.2012.11.003 Obrir en finestra nova
Projecte finançador
Saving Energy by Intelligent Monitoring
Repositori
http://hdl.handle.net/2117/19156 Obrir en finestra nova
URL
http://www.sciencedirect.com/science/article/pii/S0167865512003637 Obrir en finestra nova
Resum
The use of binary support vector machines (SVMs) in multi-classification is addressed in this paper. Margins associated to the bi-classifiers, since they depend on the geometrical disposition of the classes being separated, are, in general, of various magnitudes. In order to overcome this scaling problem, a normalization process should be applied on the SVMs’ outputs. Thus, a new normalization approach is presented based on the convex hulls that contain the classes to be separated. Furthermore...
Citació
González, L. [et al.]. A study on output normalization in multiclass SVMs. "Pattern recognition letters", 01 Febrer 2013, vol. 34, núm. 3, p. 344-348.
Paraules clau
1-v-r SVM, Convex hull, Kernel methods, Multiclassification
Grup de recerca
GREC - Grup de Recerca en Enginyeria del Coneixement
IDEAI-UPC Intelligent Data Science and Artificial Intelligence

Participants

  • Gonzalez Abril, Luis  (autor)
  • Velasco Morente, Francisco  (autor)
  • Angulo Bahón, Cecilio  (autor)
  • Ortega Ramirez, Juan Antonio  (autor)