Reproduce “Complementary Recommendation in E-commerce: Definition, Approaches, and Future Directions” 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
Linyue Li, Zhijuan Du
Li and Du provide a comprehensive survey of complementary recommendation in e-commerce, synthesizing 34 representative studies from 2009 to 2024. The work systematically defines complementary relationships, classifies various modeling approaches (content, sequence, graph-based), details solutions for complex scenarios and specific problems, and outlines critical future research directions.
Key insights
- Complementary relationships are diverse, ranging from explicit co-purchase to implicit 'use together' and 'collocation compatibility,' with a growing focus on deriving true complementarity beyond co-occurrence.
- Modeling approaches have evolved from traditional content-based and sequence-based methods to advanced Graph Neural Networks and multimodal learning for richer relationship representation.
- The field is increasingly tackling complex challenges such as asymmetric relationships, higher-order complementarity, interplay with substitution, and enhancing interpretability and personalization.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.