Part VII · Chapter 66Version 1.0 · 5 November 202635 min read
Recommendation and Ranking Systems
Key message
Recommenders allocate attention among several stakeholders and learn from the behavior they shape, so their fairness must be assessed as exposure, quality across user groups and diversity, over time and across the whole ecosystem, not as a one-time classification metric.
Five chapters or files are free. This one would be your sixth.
Leave an email address and every chapter and every PDF of the compendium opens, in this browser, for a year. Nothing else is asked, and whether we may contact you is your choice.