The MBAUM framework and its application to the ELFI system show that a combination of usage modelling (realised by applying a novel neuro-fuzzy algorithm) and user modelling, of mentalistic assumptions and behaviouroriented assumptions is possible. Moreover, this combination is beneficial in the sense that it allows for rich user-adaptivity, related to both individualisation of the user interface and personalization of system functionality. This means an important improvement over previous approaches focusing on only one aspect of adaptivity. Further developments include refinement of MBAUM components and implementations in other application systems. Special attention will be paid to forecasting of user action sequences and to adaptive help.
Davidson, B. D. & Hirsh, H. ( 1998). "Probabilistic online action prediction". AAAI Spring Symposium on Intelligent Environments.
Maes, P. ( 1994). "Agents that reduce work and information overload". Communications of the ACM, 37( 7), 31-40.
Miller, G. A.( 1956), "The magical number seven plus or minus two", Psychological Review, 63, 81-87.
Nikov, A., Delichev, S. & Stoeva, S. ( 1997). "An neuro-fuzzy approach for user interface adaptation implemented in the information system of Bulgarian parliament", In Sepälä, P. Luopajärvi, T. Nygård, C. & Mattila, M. (Eds.): Human computer-interaction, Stress and mental load, Aging, Occupational health (Vol. 1) Proc. 13th International Ergonomics Congress, (IEA'97, Tampere, Finland, June 29 - July 4, 1997), pp. 100-112. Helsinki: Finnish Institute of Occupational Health.
Pohl, W. ( 1997). "LaboUr -- Machine learning for user modelling. In Smith", M. J. et al. (Eds.): Proc. of HCI International '97, Vol. B, pp. 27-31. Amsterdam: Elsevier Science.
Pohl, W. ( 1998). "Logic-Based Representation and Reasoning for User Modelling Shell Systems". St. Augustin, Germany: infix.
Stoeva, S. & Nikov, A. ( 1999). "A fuzzy backpropagation algorithm". Fuzzy Sets and Systems, accepted for publication.