Combining Ranking with Traditional Methods for Ordinal Class Imbalance

Áreas de investigación:

• Methodology - Machine learning

Año:

2017

Tipo de publicación:

Artículo en conferencia

Palabras clave:

Ordinal classification, Class imbalance, Ranking, SVM

Autores:

• Cruz, R.
• Fernandes, K.
• Costa, J. F. Pinto
• Pérez-Ortiz, María
• Cardoso, J. S.

Volumen:

10306

Título del libro:

International Work Conference on Artificial Neural Networks (IWANN2017)

Serie:

Lecture Notes in Computer Science

Páginas:

538-548

ISBN:

978-3-319-59146-9

BibTex:

@conference{R. Cruz2017, author = "R. Cruz and K. Fernandes and J.F. Pinto Costa and Mar{\'i}a P{\'e}rez-Ortiz and J.S. Cardoso", abstract = "In classification problems, a dataset is said to be imbalanced when the distribution of the target variable is very unequal. Classes contribute unequally to the decision boundary, and special metrics are used to evaluate these datasets. In previous work, we presented pairwise ranking as a method for binary imbalanced classification, and extended to the ordinal case using weights. In this work, we extend ordinal classification using traditional balancing methods. A comparison is made against traditional and ordinal SVMs, in which the ranking adaption proposed is found to be competitive.", booktitle = "International Work Conference on Artificial Neural Networks (IWANN2017)", doi = "10.1007/978-3-319-59147-6_46", isbn = "978-3-319-59146-9", keywords = "Ordinal classification, Class imbalance, Ranking, SVM ", pages = "538-548", series = "Lecture Notes in Computer Science", title = "{C}ombining {R}anking with {T}raditional {M}ethods for {O}rdinal {C}lass {I}mbalance", url = "http://dx.doi.org/10.1007/978-3-319-59147-6_46", volume = "10306", year = "2017", }

Abstract:

In classification problems, a dataset is said to be imbalanced when the distribution of the target variable is very unequal. Classes contribute unequally to the decision boundary, and special metrics are used to evaluate these datasets. In previous work, we presented pairwise ranking as a method for binary imbalanced classification, and extended to the ordinal case using weights. In this work, we extend ordinal classification using traditional balancing methods. A comparison is made against traditional and ordinal SVMs, in which the ranking adaption proposed is found to be competitive.