Researchers have developed AlleCompanion, a new framework for improving complementary product recommendations. This system addresses the challenge of distinguishing between products that are frequently bought together and those that are genuinely complementary. By combining data-level filtering with a category-constrained Two Tower architecture and a multi-source mapping tool called ComCat, AlleCompanion effectively filters out noise from co-purchase data. The framework has been deployed at Allegro, serving over 20 million users monthly, and has shown significant increases in gross merchandise value and revenue from sponsored placements. AI
IMPACT Enhances e-commerce discovery and revenue through more accurate product pairing.
RANK_REASON This is a research paper detailing a new recommendation framework. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →