Learning How To Cook...
summary
Argues that people need DIY recommendation algorithms so users control what shapes decisions and not just corporations; shows concrete mechanics: explicit vs implicit inputs (ratings versus watch time), two process families collaborative filtering and content-based, and the common evaluation: engagement as metric which drives confirmation bias and filter bubbles. Notes existing building blocks — recommendation-as-a-service, open source libraries, and IFTTT-style grammars — and proposes practical interfaces like an ingredient list of data and influence so people can compare and choose.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:30What is decisionmaking
- 1:12Deconstructing decisionmaking
- 1:54Where are they useful
- 3:11Input
- 4:24Collaborative Filtering
- 4:57Content Based Filtering
- 5:40Recommendation Algorithms
- 8:02Engagement
- 9:59Eggbeater
- 10:30Output
- 12:45Bias
- 13:31Software
- 15:03Food
- 15:21Design
related talks
41mRecommender Systems: Crafting Effective Personalised Experiences
35mData Science for Dinner: Recommending Recipes
45mContent recommendation at iFood scale
05mDeepCook: Code Your Way Through Everyday Chores
32mLet's talk about Recommendation Systems!
19mPopping the Filter Bubble
33mPeople who liked this talk also liked ... Building Recommendation...
23mML-Powered Search & Recommendation 101