Concept Recommender

A method for improving the learning process by recommending suitable document to user on the basis of user characteristics.

Technologies used: Java, XML (JAXB), Sesame repository
Inputs: User model, current learning object being read within the presentation layer
Outputs: Recommendation of a next learning object to the user

Addressed Problems

Learners have different knowledge, interests and goals. Personalized approach to learning and education is the way, verified by long term history of education, which can assure the efficiency of learning process to each and every learner. However, in spite of progress in methods for personalized recommendation, personalization is still not that obvious in current e-learning systems.


is based to select relevant parts of the information space for the learner. Recommendation is influenced by several user-related attributes (e.g., interest, knowledge). on searching and reasoning in two mutually interconnected information spaces: knowledge items (concepts, topics) and learning objects. First, on the knowledge item level the most relevant concepts and topics for the user are selected. Afterwards, recommendation process continues on the learning object level, where selection of most suitable learning objects is accomplished based on the learning objects that are related to the most relevant knowledge items acquired in the previous step. One of advantages of our approach is that we do not need to browse the entire learning object space, but only a small subspace related to chosen knowledge items. Moreover, reasoning on the knowledge items level is more accurate as it enables better estimation of the user knowledge level.

Relevance criterions for knowledge items consist of:

Evaluation of learning objects determines a relevance of the learning objects to particular user as weighted sum of defined criterions fulfillment.Between defined criterions belong:

structure of proposed domain model.

Example of browsing knowledge item space, initialized in nodes Files, Cycles and Arrays