EduBERT: Pretrained Deep Language Models for Learning Analytics
This work addresses domain-specific NLP challenges in education for learning analytics researchers, offering incremental improvements by adapting existing methods to new data.
The paper tackled the problem of applying large pretrained language models to learning analytics tasks, showing that pre-training on student forum data improves performance beyond state-of-the-art on three text classification tasks, with a distilled version achieving best results on two tasks while reducing computational cost.
The use of large pretrained neural networks to create contextualized word embeddings has drastically improved performance on several natural language processing (NLP) tasks. These computationally expensive models have begun to be applied to domain-specific NLP tasks such as re-hospitalization prediction from clinical notes. This paper demonstrates that using large pretrained models produces excellent results on common learning analytics tasks. Pre-training deep language models using student forum data from a wide array of online courses improves performance beyond the state of the art on three text classification tasks. We also show that a smaller, distilled version of our model produces the best results on two of the three tasks while limiting computational cost. We make both models available to the research community at large.