Recommending physics exercises in moodle based on hierarchical competence profiles

Abstract

We present a prototype for an adaptive navigation system which uses a dissimilarity measure between student and exercise profiles to rank and recommend exercises. Both types of profiles are structured as a hierarchical tree. We are developing a Moodle plugin that presents the top-ranked exercises as recommendations to distance learning students. A visualization of the student competence profiles provides progress feedback within the plugin.