The users of any educational software may leave traces which concern all their activities. In collaborative learning context, these traces are very voluminous and very heterogeneous. They are the results of various interactions between the actors themselves, and between the actors and the system. Hence, first, they must be collected and filtered. Then, these traces must be analyzed to help or support the actors. In fact, the analysis process has positive effect on the role of human actors. It helps the tutor in his task of monitoring learners, the teacher (author) in his task of creating educational courses and the learner in his task of viewing his previous actions with more details. It is this context that defines our research work, which is focused on implementing a collaborative learning system based on traces called SYCATA (which is the French acronym of: SYstème pour la Collecte et l'Analyse des Traces d'Apprentissage collaboratif). This system collects all traces of actors' activities (especially learners) and groups them into five categories. It offers a multitude of forms (graphical, numerical or mixed) to show these traces to the tutors and the authors. Some traces may be viewed by learners to promote their pedagogical activities and raise their awareness. SYCATA was implemented and experimented with a sample of university students where good results were obtained.

No abstract

No abstract