Food is personal! Design and impact of personalized recommendations in e-grocery
summary
Project builds a personalized webshop for 4,000+ items using a heterogeneous network that links user and item similarities from transaction history, computes a per user–item subnetwork, fits regressions and uses the F‑test p‑value to rank products; model recalculates daily and serves category rankings. A 1.5 year effort runs an A/B test with >20,000 orders, yields a 3.3 second median reduction in shopping time and about +7% revenue. Concrete features: per‑user ranked category pages and mix of repeat‑purchase and co‑purchase signals.
AI-written summary · about
chapters (5)
related talks
27mTransform online shopping experiences with personalized recommendations
31mEverything in its Right Place: Optimising Ranking in Online Grocery
29mThe Challenges of Building a Personalized Recommendation System in E-Commerce
41mRecommender Systems: Crafting Effective Personalised Experiences
35mData Science for Dinner: Recommending Recipes
38mThe Power of the Bonus Card: Road to Personalised Search
27mOur experiments with food recommendations
32mLeveraging recommender systems to personalise search results