Taking into account the variability of the knowledge structure in Bayesian student models

  • Authors:

  • Mathieu Hibou

  • Affiliations:

  • Crip5 Université René Descartes --Paris 5, 45 rue des Saints-Pères 75270 Paris Cedex 06 France, mathieu.hibou@math-info.univ-paris5.fr

  • Venue:
  • Proceedings of the 2005 conference on Artificial Intelligence in Education: Supporting Learning through Intelligent and Socially Informed Technology
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Bayesian belief networks have been widely used in student and user modelling. Their construction is the main difficulty for their use in student modelling. The choices made about their structures (especially the arcs orientation) have consequences in terms of information circulation. The analysis we present here is that the network structure depends on the expertise level of the student. Consequently, the evolution of the network should not only be numerical (update of the probabilities) but also structural. Hence, we propose a model constituted of different networks in order to take into account these evolutions.