In this paper, we propose a machine learning based drop-out prediction process for spotting and preventing drop-out among MOOC learners early upon their interaction with the course. Two main goals are perused in this scope, first, achieving an accurate prediction at a specified instant of the course, established with the help of predictive type classifiers. Second, uncovering the underlying reasons for the predicted drop-out established with the help of explicative type classifiers. The related experimental findings show promising results.

Keywords

decision trees · machine learning · predictive models · radio frequency · testing · trajectory · tuning