Dropout Prediction over Weeks in MOOCs by Learning Representations of Clicks and Videos
This addresses dropout prediction for MOOC learners, but it is incremental as it builds on existing feature extraction techniques by adding video modeling.
The paper tackled MOOC dropout prediction by learning representations of videos and their correlation with clickstream data, resulting in statistically significant improvements in prediction accuracy.
This paper addresses a key challenge in MOOC dropout prediction, namely to build meaningful representations from clickstream data. While a variety of feature extraction techniques have been explored extensively for such purposes, to our knowledge, no prior works have explored modeling of educational content (e.g. video) and their correlation with the learner's behavior (e.g. clickstream) in this context. We bridge this gap by devising a method to learn representation for videos and the correlation between videos and clicks. The results indicate that modeling videos and their correlation with clicks bring statistically significant improvements in predicting dropout.