A predictive model for the identification of learning styles in MOOC environments

12 Oct 2019  ·  Brahim Hmedna, Ali El Mezouary, Omar Baz ·

Massive online open course (MOOC) platform generates a large amount of data, which provides many opportunities for studying the behaviors of learners. In parallel, recent advancements in machine learning techniques and big data analysis have created new opportunities for a better understanding of how learners behave and learn in environments known for their massiveness and openness. The work is about predicting learners’ learning styles based on their learning traces. The Felder Silverman learning style model (FSLSM) is adopted since it is one of the most commonly used models in technology-enhanced learning. In order to attend our objective, we analyzed data collected from the edX course “statistical learning” (session Winter 2015 and Winter 2016), administered via Stanford’s Logunita platform. The results show that decision tree performs best for all 3 dimensions, with an accuracy of higher than 98% and a reduced risk of overfitting the training data.



  Add Datasets introduced or used in this paper

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.