Multimodal mood-based annotation

The paper presents an architecture for multimodal mood-based annotation systems. The architecture aims at the implementation of interactive multimodal systems to support communities of users in the creation and management of annotations in locative media projects. The annotations are multimodal in that they can be created and accessed through visual and audio interaction. The annotations are mood-based in that they reflect the mood of the user respect to the point of interest s/he is commenting. The paper presents a definition of multimodal mood-based annotation and a description of the architecture, illustrating in particular the interaction process between users and systems through the audio interface. A concrete application of the architecture is presented: an annotative locative media project aimed at supporting tourists in creating annotations related to the Valchiavenna valley in Italy.