Content recommendation at iFood scale
summary
Challenges of building content recommendations at iFood scale are presented, focusing on increasing app personalization using contextual recommendation, feature engineering, attribute selection, experiment design, and result validation. The talk also covers strategies and algorithms used during the experimental phase, how to run large-scale production systems alongside Data Science experimentation, and the relative business-fit of different algorithms.
AI-written summary · about
chapters (4)
related talks
35mData Science for Dinner: Recommending Recipes
37mFood is personal! Design and impact of personalized recommendations in e-grocery
18mLearning How To Cook...
27mOur experiments with food recommendations
19mContent Recommendation Within Reach!
38mRecommendations at Rappi: from MVP to Data Product
50mAnalytics – Balancing ‘Advanced’ and ‘Actionable’
30mLessons from building deep learning recommendation systems