Network of Excellence Peer-to-Peer Tagged Media

EventMedia

Events are a natural way for referring to any observable occurrence grouping persons, places, times and activities. Events are also observable experiences that are often documented by people through different media. We explore this intrinsic connection between user generated media and experiences so that people can search and browse through content using a familiar event perspective.

EventMedia will provide a web-based environment that allows users to discover meaningful, surprising or entertaining connections among events by using a knowledge base of events linked to the LOD cloud, in conjunction with an event ontology (e.g. LODE). Data will be gathered from different event sources (e.g. Last.fm, Eventful, Upcoming) associated to media (e.g. Flickr, Youtube, Dailymotion) and enriched with social user profile data (e.g. Facebook, Twitter) to explore relevancy in the context of different user tasks.

 

The approach is user-driven and, having carried out explorative user studies, we are designing interfaces that support user-identified tasks while exploring the connections between people, multimedia content and events. These are the  main scientific contributions of EventMedia:

  1. User-centered concept idealization and development of a web-based environment for annotating media content with events and for exploring media collections based on the dimensions composing an event and social user profiles.
  2. Modeling of an event ontology, represented in OWL/RDF, compatible with other event models, together with a knowledge base of events and associated user generated media. Content will be obtained from large event and media directories and linked to the LOD cloud.
  3. Investigation of user interface and information visualization perspectives according to established user requirements and factual properties defined by the event ontology.

Contact: Raphaƫl Troncy; raphael.[lastname]@eurecom.fr