Reproduce “Theoretically Sound Item Similarity Measures for Recommender Systems” 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
Noah Daniëls, Joey De Pauw, Bart Goethals
The authors derive item-similarity measures from probabilistic models for implicit-feedback recommendation, where an unobserved item is unknown rather than disliked. Their multinomial model estimates candidate-item likelihoods from users’ observed histories. In held-out tests, it outperformed item-neighbor baselines on books and news, but not EASE; on movies, it offered little improvement over cosine.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.