RUO Principal

Repositorio Institucional de la Universidad de Oviedo

Ver ítem 
  •   RUO Principal
  • Producción Bibliográfica de UniOvi: RECOPILA
  • Artículos
  • Ver ítem
  •   RUO Principal
  • Producción Bibliográfica de UniOvi: RECOPILA
  • Artículos
  • Ver ítem
    • español
    • English
JavaScript is disabled for your browser. Some features of this site may not work without it.

Listar

Todo RUOComunidades y ColeccionesPor fecha de publicaciónAutoresTítulosMateriasxmlui.ArtifactBrowser.Navigation.browse_issnPerfil de autorEsta colecciónPor fecha de publicaciónAutoresTítulosMateriasxmlui.ArtifactBrowser.Navigation.browse_issn

Mi cuenta

AccederRegistro

Estadísticas

Ver Estadísticas de uso

AÑADIDO RECIENTEMENTE

Novedades
Repositorio
Cómo publicar
Recursos
FAQs

Mapping preferences into Euclidean space

Autor(es) y otros:
Luaces Rodríguez, ÓscarAutoridad Uniovi; Díez Peláez, JorgeAutoridad Uniovi; Joachims, Thorsten; Bahamonde Rionda, AntonioAutoridad Uniovi
Palabra(s) clave:

Preference learning

Matrix factorization

Graphical representations

Learning to order

Fecha de publicación:
2015-12
Editorial:

Elsevier

Versión del editor:
http://dx.doi.org/10.1016/j.eswa.2015.07.013
Citación:
Expert Systems with Applications, 42(22), p. 8588–8596 (2015); doi:10.1016/j.eswa.2015.07.013
Descripción física:
p. 8588-8596
Resumen:

Understanding and modeling human preferences is one of the key problems in applications ranging from marketing to automated recommendation. In this paper, we focus on learning and analyzing the preferences of consumers regarding food products. In particular, we explore Machine Learning methods that embed consumers and products in an Euclidean space such that their relationship to each other models consumer preferences. In addition to predicting preferences that were not explicitly stated, the Euclidean embedding enables visualization and clustering to understand the overall structure of a population of consumers and their preferences regarding the set of products. Notice that consumers’ clusters are market segments, and products clusters can be seen as groups of similar items with respect to consumer tastes. We explore two types of Euclidean embedding of preferences, one based on inner products and other based on distances. Using a real world dataset about consumers of beef meat, we find that both embeddings produce more accurate models than a tensorial approach that uses a SVM to learn preferences. The reason is that the number of parameters to learned in embeddings can be considerably lower than in the tensorial approach. Furthermore, we demonstrate that the visualization of the learned embeddings provides interesting insights into the structure of the consumer and product space, and that it provides a method for qualitatively explaining consumer preferences. In addition, it is important to emphasize that the approach presented here is flexible enough to allow its use with different levels of knowledge about consumers or products; therefore the application field is very wide to grasp an accurate understanding of consumers’ preferences

Understanding and modeling human preferences is one of the key problems in applications ranging from marketing to automated recommendation. In this paper, we focus on learning and analyzing the preferences of consumers regarding food products. In particular, we explore Machine Learning methods that embed consumers and products in an Euclidean space such that their relationship to each other models consumer preferences. In addition to predicting preferences that were not explicitly stated, the Euclidean embedding enables visualization and clustering to understand the overall structure of a population of consumers and their preferences regarding the set of products. Notice that consumers’ clusters are market segments, and products clusters can be seen as groups of similar items with respect to consumer tastes. We explore two types of Euclidean embedding of preferences, one based on inner products and other based on distances. Using a real world dataset about consumers of beef meat, we find that both embeddings produce more accurate models than a tensorial approach that uses a SVM to learn preferences. The reason is that the number of parameters to learned in embeddings can be considerably lower than in the tensorial approach. Furthermore, we demonstrate that the visualization of the learned embeddings provides interesting insights into the structure of the consumer and product space, and that it provides a method for qualitatively explaining consumer preferences. In addition, it is important to emphasize that the approach presented here is flexible enough to allow its use with different levels of knowledge about consumers or products; therefore the application field is very wide to grasp an accurate understanding of consumers’ preferences

URI:
http://hdl.handle.net/10651/35738
ISSN:
0957-4174
DOI:
10.1016/j.eswa.2015.07.013
Patrocinado por:

The research reported here is supported in part under Grant TIN2011-23558 from the Ministerio de Economía y Competitividad, Spain. The paper was written while A. Bahamonde was visiting Cornell University with Grants Uniovi-Feb-2012 and Uniovi-Mar-2012 from Movilidad Campus de Excelencia Internacional (Universidad de Oviedo) and BOE-A-2013-4383 from Programa Nacional de Movilidad de Recursos Humanos (Ministerio de Educación, Cultura y Deporte, Spain)

Colecciones
  • Artículos [37532]
  • Informática [872]
  • Investigaciones y Documentos OpenAIRE [8365]
Ficheros en el ítem
Thumbnail
untranslated
Postprint (1.542Mb)
Métricas
Compartir
Exportar a Mendeley
Estadísticas de uso
Estadísticas de uso
Metadatos
Mostrar el registro completo del ítem
Página principal Uniovi

Biblioteca

Contacto

Facebook Universidad de OviedoTwitter Universidad de Oviedo
El contenido del Repositorio, a menos que se indique lo contrario, está protegido con una licencia Creative Commons: Attribution-NonCommercial-NoDerivatives 4.0 Internacional
Creative Commons Image