EDU-EX, an authoring tool for generating intelligent tutoring systems (ITSs), is presented in this paper. With this tool, the course contents are dynamically adapted, in real time, to the profile of the student. The intended audience for this paper is researchers studying auto-regulated intelligent tutoring systems.

The tool is realized by separating information on the subject from information on the inference mechanism, and from information of a pedagogical nature.

The knowledge base is represented by objects, and by a list of properties associated with each object, the values of which define the state. The properties allow the establishment of relationships between objects, determine the ways in which the system operates, and describe the states through which the system passes. The reasoning process manages and processes information in the knowledge base, depending on the information obtained by the student, and taking into account the information contained within the knowledge about the objects and their properties.

Because it is visually oriented, EDU-EX allows users to develop ITSs independently on any subject, regardless of whether they are professionals or people without specific background in IT. Moreover, EDU-EX assists developers by checking, in real time, that the database is complete, and that no compulsory object or property is left undefined. Finally, an ITS so generated can be easily modified.

The results of an experiment using a course on Windows 3.1, developed using EDU-EX, are analyzed.

The paper is well organized, but lacks technical information about the system’s implementation. The bibliographical references and an analysis of ITS authoring tools will give the reader an idea of the literature in the field.