Toward Automatic User Profiling for Personalized Acces to Information within Collaborative Learning System

The use of modern educational technology methods, in order to support learning as well as collaboration, has become an important area of research. Specially with the rise of the internet and web 2.0 platforms that have transformed users’ role from mere content consumers to fully content consumers-producers. Furthermore, people engaged in collaborative learning capitalize on one another’s resources and skills, unlike individual learning. This paper proceeds with a categorization of the main tools and functions that characterise personalization learning aspect, in order to discuss their trade-offs with collaborative learning systems. It proposes a framework of a personalized information research (IR) within a collaborative learning system, incorporating the characterization of the research type carried by the query, as well as modeling and constructing semantic users’ profiles. We use the context of the user query into a prediction mechanism of search type, based on a previous identification of users’ levels and interests. The paper concludes by presenting an experiment results, revealing that the use of the subject ontology extension approach, satisfyingly contribute to an improvement in the accuracy of system’s recommendations.