csh_logo

Ranks of Disparity

Who gets ranked first—and who gets left out?

This interactive website brings hyperFA*IR to life: a framework for understanding and assessing whether rankings are fair with respect to protected attributes—groups often at a disadvantage, such as women in certain gender-based comparisons—and how they can be improved.

The website has two main parts:

  • The homepage story introduces the core concepts of ranking fairness, shows how bias can emerge, and explains how to assess and improve rankings.
  • The explorer page offers toy cases from real-world scenarios like university admissions, job hiring, and scholarships. It also provides a playground, where you can adjust parameters to better understand how fairness and ranking outcomes interact. In addition, you can upload your own dataset to assess the fairness of an existing ranking and explore possible improvements. Results from both playground and your own data can be downloaded for further use.

Publication

Mauritz N. Cartier van Dissel, Samuel Martin-Gutierrez, Lisette Espín-Noboa, Ana María Jaramillo, and Fariba Karimi. 2025. hyperFA*IR: A hypergeometric approach to fair rankings with finite candidate pool. doi.org/10.1145/3715275.37321

• Code for reproducing the results: Github

Credits

We thank DiceBear for providing the free avatar icons used throughout this visualization. The avatars are generated using the DiceBear API and are licensed under CC0 1.0.

csh_logo
More Visualizations
CSH Visuals