Graph theory based model for learning path recommendation

• We propose a model to build a learning design recommendation system.
• The presented model is based on graph theory applied to large repositories of learning objects.
• From this model, we propose, implement and test an approach using the concept of cliques to recommend learning paths.

Learning design, the activity of designing a learning path, can be a complex task, especially for learners. A learning design recommendation system would help self-learners find appropriate learning objects and build efficient learning paths during their learning journey. Educational Data Mining (EDM) has provided an impressive amount of novelties related to learning object recommendation systems. However, most of the solutions proposed thus far do not take into account eventual competency dependencies among learning objects and/or are not designed for large repositories of interdependent learning objects. We propose a model to build a learning design recommendation system based on graph theory. From this model, we propose, implement and test an approach using the concept of cliques to recommend learning paths.