English español
Búsqueda
 

Repositorio de la Universidad de Oviedo > Producción Bibliográfica de UniOvi: RECOPILA > Artículos >

Use este identificador para citar o enlazar este ítem: http://hdl.handle.net/10651/30616

Título : Binary relevance efficacy for multilabel classification
Autor(es) y otros: Luaces Rodríguez, Óscar
Díez Peláez, Jorge
Barranquero Tolosa, José
Coz Velasco, Juan José del
Bahamonde Rionda, Antonio
Palabras clave: Multilabel classification
Binary relevance
Synthetic datasets
Label dependency
Fecha de publicación : 2012
Editorial : Springer
Versión del editor: http://dx.doi.org/10.1007/s13748-012-0030-x
Citación : Progress in Artificial Intelligence, 1(4), p. 303-313 (2012); doi:10.1007/s13748-012-0030-x
Descripción física: p. 303-313
Resumen : The goal of multilabel (ML) classi cation is to induce models able to tag objects with the labels that better describe them. The main baseline for ML classi- cation is Binary Relevance (BR), which is commonly criticized in the literature because of its label independence assumption. Despite this fact, this paper discusses some interesting properties of BR, mainly that it produces optimal models for several ML loss functions. Additionally, we present an analytical study about ML benchmarks datasets, pointing out some shortcomings. As a result, this paper proposes the use of synthetic datasets to better analyze the behavior of ML methods in domains with di erent characteristics. To support this claim, we perform some experiments using synthetic data proving the competitive performance of BR with respect to a more complex method in di cult problems with many labels, a conclusion which was not stated by previous studies
URI : http://hdl.handle.net/10651/30616
ISSN : 2192-6352
2192-6360
Aparece en las colecciones: Artículos
Informática
Investigaciones y Documentos OpenAIRE

Ficheros en este ítem:

Fichero Descripción Tamaño Formato
Binary Relevance Efficacy for Multilabel Classification.pdfPostprint1,18 MBAdobe PDFVisualizar/Abrir


Exportar a Mendeley


Los ítems de DSpace están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.

 

Base de Datos de Autoridades Biblioteca Universitaria Consultas / Sugerencias