Revision History
Predictive Parity · 7 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the false claim that calibration implies predictive parity, replaced a composite Kleinberg-Mullainathan-Raghavan quote with their actual wording, removed 'test fairness' as a synonym (it means calibration), corrected the ProPublica sample behind the COMPAS error rates (7,214, not 6,172 defendants), and fixed the two-feature model accuracy in Dressel and Farid (66.8%, not 66.6%) (plus 6 minor fixes)
Version 7 · Nov 4, 2026, 11:14 PM
- Improve internal links (2026-10-13)
Version 6 · Oct 13, 2026, 10:05 AM
- Added 6 contextual internal links
Version 5 · Jul 23, 2026, 03:02 PM