Reproduce “Predicting Engagement in Video Lectures” 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
Sahan Bulathwela, María Pérez-Ortiz, Aldo Lipani, Emine Yilmaz, John Shawe-Taylor
The authors estimate population-level watch engagement for lectures from resource features, so rankings can be produced before learner feedback is available. A random forest using content and video-specific features reaches pairwise accuracy of .744 ± .011, compared with .724 ± .014 using content features alone, in five-fold cross-validation. This predicts a watch-time proxy for peer-reviewed conference lectures, not learning gains or individual preferences.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.