Discovering internal social relationship for influence-aware service recommendation
Existing approaches, such as semantic content-based or Collaborative Filtering-based recommendations, fail to exploit social aspects of services because services lack social relationships and do not consider social influence. In this paper, we propose a methodology for connecting distributed service...
Gespeichert in:
Veröffentlicht in: | Multimedia tools and applications 2017-09, Vol.76 (18), p.18193-18220 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Existing approaches, such as semantic content-based or Collaborative Filtering-based recommendations, fail to exploit social aspects of services because services lack social relationships and do not consider social influence. In this paper, we propose a methodology for connecting distributed services in a global social service network (GSSN) to facilitate discovering internal social relationship for social influence-aware service recommendation. First, we propose a novel platform for constructing a GSSN by linking distributed services with social links based on quality of social link. We then propose a flexible model of the effective awareness of social influence, which provides a quantitative measure of the strength of influence between services. Next, a novel social influence-aware service recommendation approach is proposed based on GSSN using internal social relationship among services. The experimental results demonstrated that our new approach can solve the service recommendation problem with a low usage threshold and high accuracy, where the user preferences are exploited by a recommend-as-you-go method. |
---|---|
ISSN: | 1380-7501 1573-7721 |
DOI: | 10.1007/s11042-016-3437-8 |