Reproduce “A Toolbox for Modelling Engagement with Educational Videos” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuxiang Qiu, Karim Djemili, Denis Elezi, Aaneel Shalman, María Pérez-Ortiz, Emine Yilmaz, John Shawe-Taylor, Sahan Bulathwela
TrueLearn and PEEKC support prediction of engagement with educational videos by linking video fragments to Wikipedia concepts and updating learner states from viewing interactions. The evaluation predicts whether a learner watches at least 75% of the next fragment. On the AI/ML-heavy PEEKC dataset, TrueLearn Novelty reaches 65.53% F1 versus 61.22% for the strongest baseline, measuring watch time rather than learning.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.