Semantic Composition of Lecture Subparts for a Personalized e-Learning

In this paper we propose an algorithm for personalized learning based on a user's query and a repository of lecture subparts --i.e., learning objects-- both are described in a subset of OWL-DL. It works in two steps. First, it retrieves lecture subparts that cover as much as possible the user's query. The solution is based on the concept covering problem for which we present a modified algorithm. Second, an appropriate sequence of lecture subparts is generated. Indeed, the different lecture subparts are only reachable when a given prerequisite is fulfilled, i.e., the learner must have a minimal background knowledge to be able to assimilate the requested learning object. Therefore, our algorithm takes into account the user's knowledge to generate a personalized lecture composition and suggests a flow of learning objects to the user.