De la modélisation sémantique des événements vers l'enrichissement et la recommandation

Khrouf, Houda; Troncy, Raphaël
Revue d'Intelligence Artificielle, N°2-3, 2014

Many sites have recently witnessed a rapid growth providing information about past
and upcoming events, of which some may display media captured at these events. This information is, however, often incomplete and always locked into the sites. Our goal is to mine in real time the connection between these distributed datasets using semantic web technologies. This is a key advantage to ensure a better information completeness and to deliver enriched views of events. Moreover, thousands of events are daily created which require a powerful recommender system in order to enhance the user experience. We propose a new recommendation approach leveraging the semantic description of events as well as social information. Finally, we present a web application that aims to meet the user needs: relive and discover experiences based on media, and support decision making for attending upcoming events.

Type:
Journal
Date:
2014-09-01
Department:
Data Science
Eurecom Ref:
4407
Copyright:
Lavoisier

PERMALINK : https://www.eurecom.fr/publication/4407