Multimedia Content Personalization Based on Peer-Level Annotation

In this paper, we propose an architecture which supports metadata extraction by exploring interaction mechanisms among users and content. The interaction activity addressed in this work is related to peer-level annotation, where any user acts as author, being able to enrich the content by making annotations, using, for instance, pen-based devices. Peer-level annotation makes users comfortable when taking digital notes, as they do in every day life. This is an advantage over hierarchical authoring, which is a time-consuming task usually employed by content providers. The set of metadata extracted by peer-level annotation is used in this paper to generate a user's profile, which contains preferences about the content, and may be further explored by personalization services.