Artículo
A study on output normalization in multiclass SVMs
Autor/es | González Abril, Luis
Velasco Morente, Francisco Angulo, C. Ortega Ramírez, Juan Antonio |
Departamento | Universidad de Sevilla. Departamento de Lenguajes y Sistemas Informáticos Universidad de Sevilla. Departamento de Economía Aplicada I |
Fecha de publicación | 2013-02 |
Fecha de depósito | 2023-02-14 |
Publicado en |
|
Resumen | 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, ... 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, a theoretical study is developed which justifies the proposed approach, and an interpretation is provided. An empirical study is also carried out to compare this normalization with others found in the literature. |
Agencias financiadoras | Ministerio de Economía, Innovación y Ciencia |
Identificador del proyecto | Simon-TIC-8052 |
Cita | González Abril, L., Velasco Morente, F., Angulo, C. y Ortega Ramírez, J.A. (2013). A study on output normalization in multiclass SVMs. Pattern Recognition Letters, 34 (3), 344-348. https://doi.org/10.1016/j.patrec.2012.11.003. |
Ficheros | Tamaño | Formato | Ver | Descripción |
---|---|---|---|---|
A study on output normalization ... | 203.5Kb | [PDF] | Ver/ | |