Abstract—Motivated with the goal to provide dynamic assembly and personalization of learning content parts, we propose an ontology-based solution implemented as an integrated learning environment called TANGRAM. TANGRAM relies on two ontologies for representing learning object (LO) content structure and LO content type (i.e. pedagogical role). LO content described by those two ontologies is further annotated with concepts of a domain ontology, while a learning paths ontology is used to specify pedagogical relations (e.g. prerequisites) among domain concepts. A user model ontology is defined to represent relevant information about TANGRAM’s users. The paper presents the employed ontologies, in the context of user modeling and personalization. Furthermore, it describes the algorithm we defined to dynamically assemble content units into learning content personalized to the user’s domain knowledge, preferences, and learning styles. We also discuss our experiences with dynamic content generation and summarize results of the conducted evaluation study. Although TAGRAM is a general-purpose learning environment, in this paper, we analyze it in the domain of intelligent information systems.

Index Terms—ontologies, personalized learning, semantic annotation, dynamic content assembly

Cite: Jelena Jovanovic, Dragan Gaševic, and Vladan Devedžic, "TANGRAM for Personalized Learning Using the Semantic Web Technologies," Journal of Emerging Technologies in Web Intelligence, Vol. 1, No. 1, pp. 6-21, August 2009.