Incremental learning of user models - an experimental testbed

The next generation of consumer goods, including computers, will be much more sophisticated in order to cope with a less technologically literate user base. A user model is an essential component for “user friendliness”, enabling the behaviour of a system to be tailored to the needs of a particular user. In this paper, we briefly outline the problems of user modelling, especially uncertainty, understandability, adaptability and the difficulty of obtaining data. A new user modelling approach (FILUM) is described, using inheritance from prototypes which are implemented as support logic programs. An experimental testbed based on the iterated prisoner’s dilemma, which allows the generation of unlimited data for learning or testing, is proposed. The system is implemented using Fril++, an object-based logic programming language with uncertainty.