Reproduce “Graph Convolution Machine for Context-aware Recommender System” 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
Jiancan Wu, Xiangnan He, Xiang Wang, Qifan Wang, Weijian Chen, Jianxun Lian, Xing Xie
GCM develops a context-aware recommender that connects users and items through interaction edges, using each edge’s context in graph propagation before an FM decoder scores candidates. In offline tests that hold out each user’s last interaction, GCM outperforms tested baselines across three datasets; on Yelp-NC, HR@10 is 0.1046 versus GIN’s 0.0866. The results do not establish dynamic preference modeling or universal gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.