The project investigates issues concerning the construction and maintenance of declarative knowledge bases for Companion-systems. It pools the symbolic modelling of all information the planning-, decision-making-, interaction-, and dialogue-components of such systems need to rely upon.
The knowledge base of a Companion-system consists of various components: the planning model represents goals and courses of action for both the user and the system; the user model contains well-structured user categories and profiles, which are used to individualize the system’s functionality; the knowledge model holds the ontology of the application domain; and the world model provides the current overall situation.
Situation information is composed of information about the environment and the user including his/her current emotional state. As this information is based on sensory input, advanced techniques to integrate symbolic and subsymbolic information processing will be developed. In this context, dealing with uncertainty in a consistent way is of particular importance.