Applying Recent Innovations from NLP to MOOC Student Course Trajectory Modeling
This work addresses course trajectory prediction for MOOC students, but appears incremental as it adapts existing NLP methods without new breakthroughs.
This paper tackles improving predictive methods for MOOC student course trajectory modeling by applying NLP innovations, investigating LSTM networks with regularization and Transformer architectures, but does not report concrete numerical results.
This paper presents several strategies that can improve neural network-based predictive methods for MOOC student course trajectory modeling, applying multiple ideas previously applied to tackle NLP (Natural Language Processing) tasks. In particular, this paper investigates LSTM networks enhanced with two forms of regularization, along with the more recently introduced Transformer architecture.