Cross-cultural contextualisation for recommender systems

Hong, Minsung and An, Sojung and Akerkar, Rajendra and Camacho Fernandez, David and J. Jang, Jason (2019). Cross-cultural contextualisation for recommender systems. "Journal of Ambient Intelligence And Humanized Computing" ; pp. 1-12. ISSN 1868-5137. https://doi.org/10.1007/s12652-019-01479-9.

Description

Title: Cross-cultural contextualisation for recommender systems
Author/s:
  • Hong, Minsung
  • An, Sojung
  • Akerkar, Rajendra
  • Camacho Fernandez, David
  • J. Jang, Jason
Item Type: Article
Título de Revista/Publicación: Journal of Ambient Intelligence And Humanized Computing
Date: September 2019
ISSN: 1868-5137
Volume: 0
Subjects:
Freetext Keywords: Cross-cultural contextualisation; Computational analysis; Recommender system; Matrix factorisation; Cultural analysis; Smart cultural heritage
Faculty: E.T.S.I. de Sistemas Informáticos (UPM)
Department: Sistemas Informáticos
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img] PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (11MB)

Abstract

Cultural Heritage (CH) domain is rapidly moving from traditional heritage sites into smart cultural heritage environment through various technologies. As one of the important technologies in the smart space, Recommender Systems (RSs) have been widely utilised to personalised services and matching visitors? goals and behaviours. Whereas, cultural difference is often considered a barrier to technology transfer or adoption. However, few studies focus on how the cultural factor influences recommendation despite cultural difference largely affects user preferences in the RSs. Furthermore, existing researches have mainly analysed evaluation results of their recommendation to reveal cultural differences, rather than utilising the cross-cultural factors into RSs. In this paper, we propose a novel concept of cross-cultural contextualisation and a model to compute the cross-cultural factor affecting users (countries or cultures) preferences by using matrix factorisation and clustering techniques. In addition, we discuss how to apply the model to RSs in CH domain through cross-domain recommendation techniques. Note that the two computational techniques were used to analyse cross-cultural factors which impact to user preferences, rather than to recommend items. In other words, the proposed model and computing results capable of utilisation into the other RSs as well as various research fields. Results of experiments with a real-world dataset showed effectiveness of the proposed model and supported that there is cultural difference influencing users? rating behaviours. Furthermore, a systematic analysis of dataset and the experimental results presented that individual users could be considered as country-wise groups for the model to analyse the cross-cultural factors.

More information

Item ID: 67829
DC Identifier: https://oa.upm.es/67829/
OAI Identifier: oai:oa.upm.es:67829
DOI: 10.1007/s12652-019-01479-9
Official URL: https://link.springer.com/article/10.1007/s12652-019-01479-9
Deposited by: Memoria Investigacion
Deposited on: 22 Feb 2022 16:35
Last Modified: 22 Feb 2022 16:35
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM