Explaining and indexing solutions for physics learning

• El Hore, Ali ^[2] ; Tazi, Said ^[1]
  1. [1] Laboratory for Analysis and Architecture of Systems

     Laboratory for Analysis and Architecture of Systems

     Arrondissement de Toulouse, Francia

     
  2. [2] Université Toulouse I
• Localización: Proceedings of the IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2005): Porto, Portugal, December 14-16, 2005 / coord. por Kinshuk Kinshuk, Demetrios Sampson, Pedro Isaías, Luís Rodrigues, Patrícia Barbosa, Vol. 2, 2005 (Short Papers), ISBN 972-8924-05-4, págs. 532-535
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • The basic hypothesis of the ongoing research presented in this paper is that the observation of problem solving by students facilitates the learning process. The externalization of the problem solving process may help to facilitate this observation. This work has been developed with context of the Pero system which is an automatic solver of exercises in physics. This paper focuses on two aspects of this system: the explanation and the indexation. Both aspects are based on an original model. This model is based on the notion of intention of the solving process. The solving process has a graph like stricture where nodes correspond to the solving states and the arcs to the actions the system. This model assumes that any action made by system Pero has its own intention. The intention is defined by the goal, the means and the reason for the action. The explanation of an exercise solution is a sequence of intentions of all actions performed by the Pero to provide a proof for the exercise and for each exercise solved by Pero is indexed by your concepts (reasons) used in its solution. This paper shows how this intentional structure eases the explanation and the indexation at run time. For the validation of our approach, a complete application on exercises in electricity was established.