Concept
Recommendation system
You might say: the algorithm, For You page, browse features
The software that picks which videos each viewer sees. YouTube describes it as following the audience: it learns from what viewers choose, watch and enjoy. A first stage picks hundreds of candidates from millions; a second ranks them.
- What it does
- Brings most views on big platforms, so it multiplies whatever viewers respond to.
- Ask
- What did viewers who saw this do next: click, watch, finish, come back?
- Trouble
- You optimise for a guessed “algorithm” instead of for viewers.
Connections
- Raises: Impressions
Sources
- YouTube Blog, On YouTube's recommendation system, What is a recommendation system?; How we personalize recommendations
Recommendations drive a significant amount of the overall viewership on YouTube, even more than channel subscriptions or search.
- YouTube Help, How YouTube recommendations work, YouTube's recommendation system
Our recommendation system is constantly evolving, learning every day from over 80 billion pieces of information we call signals.
- Paul Covington, Jay Adams and Emre Sargin, Deep Neural Networks for YouTube Recommendations, Abstract
The paper is split according to the classic two-stage information retrieval dichotomy: first, we detail a deep candidate generation model and then describe a separate deep ranking model.
- TikTok Newsroom, How TikTok recommends videos #ForYou, What factors contribute to For You?
This feed is powered by a recommendation system that delivers content to each user that is likely to be of interest to that particular user.
Use it on your project
teardown-apply: hold "Recommendation system" (recommendation-system) from the teardown at https://teardownatlas.com/content-creation/ against my project
Paste it into a coding agent that has the teardown skills.
An AI agent wrote this map with the teardown skill. It can contain errors: check the sources before you rely on a part. About and legal