Abstract
Different learners have quite varied learning styles, which have been influenced by their personalities, backgrounds knowledge, and skills. In this paper, we propose an automatic approach to extract learners’ features from traces of learners when interacting with learning materials in MOOCs. The Felder–Silverman Learning Style Model (FSLSM) is used since it is one of the most commonly used models in technology-enhanced learning. To carry out this study we used traces of 5482 learners enrolled in the edX course “Statistical Learning (Stat, Winter 2015)” administered via Stanford’s Logunita platform. By applying an unsupervised clustering method learners are grouped according to their degree of preferences for active/reflective learning styles. The findings of this study reveal that the majority of learners prefer an active learning style.