Automating the analysis of problem-solving activities in learning environments: the co-lab case study

Abstract

The analysis of problem-solving activities carried out by students in learning settings involves studying the students' actions and assessing the solutions they have created. This analysis constitutes an ideal starting point to support an automatic intervention in the student activity by means of feedback or other means to help students build their own knowledge. In this paper, we present a model-driven framework to facilitate the automation of this problemsolving analysis and of providing feedback. This framework includes a set of authoring tools that enable software developers to specify the analysis process and its intervention mechanisms by means of visual languages. The models specified in this way are computed by the framework in order to create technological support to automate the problem-solving analysis. The use of the framework is illustrated thanks to a case study in the field of System Dynamics where problem-solving practices are analysed.