To design an adaptive intelligent tutoring system which can manage both different disciplinary domains and a guide for the learner is difficult. The specialization of the analysis treatments is responsible for the loss of reusability in other disciplinary domains. The analysis is didactic and thus strongly connected to the domain concerned. It results that an intelligent tutoring system is consequently, specialized in a type of taught knowledge and not easily transposable to other domains. To propose a model transposable to different domains of learning, the former has to take into account this diversity and to situate the learning activity. In this paper, we will show how to produce a guide model parameterized by the learning domain. Our objective was to develop an adaptive intelligent tutoring system based on item response theory and metrics, adapted for letting the learners work in several disciplinary fields in the University of Annaba. In this context, our constraint is threefold: to represent knowledge relative to several disciplinary domains, to propose interactive activities to the learners and finally, to be able to support student guidance in her/his course by proposing her/him relevant support activities when he meets difficulties.