Competency-based personalization for massive online learning

This paper investigates the problem of personalization in massive open online courses (MOOC) based on a target competency profile and a learning scenario model built for the course. To use such a profile for adaptive learning and resource recommendation, we need to be able to compare competencies to help match the competencies of learners with those involved in other learning scenario components (actors, activities, resources). We present a method for computing relations between competencies based on a structured competency model. We use this method to define recommendation agents added to a MOOC learning scenario. This approach for competency comparison has been implemented within an experimental platform called TELOS. We propose to integrate these functionalities to a MOOC platform such as Open-edX. We present a personalization process and we discuss the tools needed to implement the process.