Nowadays, Ubiquitous Computing allows a high number of multime¬dia contents to be accessible anywhere and anytime by using several devices, also characterized from limited computational and storage re¬sources. To support users in multimedia choices, different recommender systems have been proposed in the past, but any of them considers the effects of the exploited devices, even though users show different behaviours in presence of different devices. This paper tries to give a contribution in this setting, proposing a new agent-based recommender system in which each device is provided with a client agent able to monitor the user’s behaviour performed on that device. A unique server agent associated with that user collects from his/her de¬vices this information to build a global profile, periodically returned to the client agents. Finally, recommendations of multimedia resources are generated from the collaboration among a recommender agent, associated with a Web site, and the client agent running on the device currently exploited by the user. Some experiments confirm the high quality of the recommendations generated by the proposed approach.

An Agent-Based Approach to Adapt Multimedia Web Content in Ubiquitous Environment

ROSACI D;SARNE' G
2011-01-01

Abstract

Nowadays, Ubiquitous Computing allows a high number of multime¬dia contents to be accessible anywhere and anytime by using several devices, also characterized from limited computational and storage re¬sources. To support users in multimedia choices, different recommender systems have been proposed in the past, but any of them considers the effects of the exploited devices, even though users show different behaviours in presence of different devices. This paper tries to give a contribution in this setting, proposing a new agent-based recommender system in which each device is provided with a client agent able to monitor the user’s behaviour performed on that device. A unique server agent associated with that user collects from his/her de¬vices this information to build a global profile, periodically returned to the client agents. Finally, recommendations of multimedia resources are generated from the collaboration among a recommender agent, associated with a Web site, and the client agent running on the device currently exploited by the user. Some experiments confirm the high quality of the recommendations generated by the proposed approach.
2011
9781613501078
Web multimedia Content; Recommender systems; Multi-agent systems
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12318/10034
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 7
  • ???jsp.display-item.citation.isi??? ND
social impact