One-Sided Prototype Selection on Class Imbalanced Dissimilarity Matrices
Ver/ Abrir
Impacto
Scholar |
Otros documentos de la autoría: Millán Giraldo, Mónica; García, Vicente; Sánchez Garreta, Josep Salvador
Metadatos
Mostrar el registro completo del ítemcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7038
comunitat-uji-handle3:10234/54899
comunitat-uji-handle4:
INVESTIGACIONEste recurso está restringido
http://dx.doi.org/10.1007/978-3-642-34166-3_43 |
Metadatos
Título
One-Sided Prototype Selection on Class Imbalanced Dissimilarity MatricesFecha de publicación
2012Editor
Springer Berlin HeidelbergISBN
978-3-642-34165-6ISSN
0302-9743; 1611-3349Cita bibliográfica
Millán Giraldo, Mónica; García, Vicente ; Sánchez Garreta, José Salvador. "One-Sided Prototype Selection on Class Imbalanced Dissimilarity Matrices". En: Structural, Syntactic, and Statistical Pattern Recognition – Joint IAPR International Workshop, SSPR&SPR 2012, Hiroshima, Japan, November 7-9, 2012. Proceedings / Gimel’farb, Georgy [et al.] (Eds.). Berlin : Springer, 2012. (Lecture Notes in Computer Science; 7626) . ISBN: 978-3-642-34165-6. pp. 391-399Tipo de documento
info:eu-repo/semantics/bookPartVersión de la editorial
http://link.springer.com/chapter/10.1007%2F978-3-642-34166-3_43Palabras clave / Materias
Resumen
In the dissimilarity representation paradigm, several prototype selection methods have been used to cope with the topic of how to select a small representation set for generating a low-dimensional dissimilarity space. ... [+]
In the dissimilarity representation paradigm, several prototype selection methods have been used to cope with the topic of how to select a small representation set for generating a low-dimensional dissimilarity space. In addition, these methods have also been used to reduce the size of the dissimilarity matrix. However, these approaches assume a relatively balanced class distribution, which is grossly violated in many real-life problems. Often, the ratios of prior probabilities between classes are extremely skewed. In this paper, we study the use of renowned prototype selection methods adapted to the case of learning from an imbalanced dissimilarity matrix. More specifically, we propose the use of these methods to under-sample the majority class in the dissimilarity space. The experimental results demonstrate that the one-sided selection strategy performs better than the classical prototype selection methods applied over all classes. [-]
Derechos de acceso
© Springer, Part of Springer Science+Business Media
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/restrictedAccess