Recently, the two research areas of Intelligent Interactive Information Presentation and Natural and Multimodal Interactivity have emerged as key requirements for the usability and effectiveness of information systems.

• Intelligent Interactive Information Presentation relates to the ability of a computer system to automatically produce multimodal information presentations, taking into account the specifics about the user, such as needs, interests, or knowledge, and engaging in a collaborative interaction that helps the retrieval of relevant information and its understanding on the part of the user. In order to build intelligent multimodal interactive presentation systems, different technologies must be exploited synergistically, such as natural language generation, language-based interaction and advanced user modelling. Also, the media and modalities used impose further constraints. For instance, the same meaning can be conveyed differently according to the communication channels available for the specific application setting. In particular, when several media and modalities are involved, presentations require their appropriate co-ordination.

   

• Natural and Multimodal Interactivity has the goal of ultimately supporting people in communicating with an information system in the same ways in which they would communicate with one another. This requires not only natural output presentation but also that users are able to present input to the computer in ways which feel natural to them, e.g. by talking to the computer whilst also pointing with the hand/arm and wrinkling the eyebrows, all of which is being understood by the system. The construction of natural interactive and multimodal systems requires equal focus on input and output so that technologies such as natural language understanding, dialogue management, speech and other signal processing techniques are needed in addition to those required for output presentation. Moreover, the issue of choosing the appropriate media and modalities for the application at hand increases in complexity when both input and output must be taken into account.

Workshop Goal

The aim of the workshop is to bring together researchers from the fields of Natural Language Generation, Conversational Spoken Dialogue, Multimodal Communication, Life-like Characters, Adaptive Hypermedia, User Modelling, and technologists interested in new media and modalities. Paper authors will present innovative results or introduce challenging issues in at least one of the two areas mentioned above in order to stimulate discussion. The workshop will include short presentations and brainstorming sessions initiated by invited speakers

The workshop is sponsored by the CLASS Project, funded by the Human Language Technologies (HLT), which is part of the Information Society Technologies Programme of the EC's 5th Framework.

CLASS (http://www.class-tech.org) was created on the request of the European Commission for the purpose of stimulating cross-project collaborations among HLT projects and between HLT projects and relevant projects world-wide, and for suggesting how the Commission can better support and meet the needs of advanced HLT research projects. Researchers working on European projects are particularly encouraged to participate in the workshop.

Workshop Program & Online Proceedings

Workshop Co-chairs

Oliviero Stock, ITC-irst, Italy

Program Committee

Elisabeth André, University of Augsburg, Germany