Adapting the CMIM algorithm for multi-label feature selection. A comparison with existing methods

The multi-label paradigm has recently attracted the attention of the machine learning community, multi-label problems being those which do not have only one class but several binomial classes instead. Although intensive research has been carried on lately into the multi-label classification paradigm...

Descripción completa

Detalles Bibliográficos
Autores: Bermejo López, Pablo, Gámez Martín, José Antonio, Puerta Callejón, José Miguel
Tipo de recurso: artículo
Fecha de publicación:2018
País:España
Institución:Universidad de Castilla-La Mancha
Repositorio:RUIdeRA. Repositorio Institucional de la UCLM
OAI Identifier:oai:ruidera.uclm.es:10578/28304
Acceso en línea:http://hdl.handle.net/10578/28304
Access Level:acceso abierto
Palabra clave:Adaptation
Feature subset selection
Multilabel
Descripción
Sumario:The multi-label paradigm has recently attracted the attention of the machine learning community, multi-label problems being those which do not have only one class but several binomial classes instead. Although intensive research has been carried on lately into the multi-label classification paradigm, this is not the case of feature subset selection methods. In this work we propose an adaptation of the well-known CMIM feature selection algorithm, which is capable of approximating the conditional multivariate mutual information of each candidate attribute with respect to the whole set of labels. This capacity to search any degree of interaction among labels, is the reason why our proposal performs better than other state-of-the-art algorithms when the dataset on which it is run contains correlated labels.