FairSight: Visual Analytics for Fairness in Decision Making

  • 2019-08-01 01:59:54
  • Yongsu Ahn, Yu-Ru Lin
  • 24

Abstract

Data-driven decision making related to individuals has become increasinglypervasive, but the issue concerning the potential discrimination has beenraised by recent studies. In response, researchers have made efforts to proposeand implement fairness measures and algorithms, but those efforts have not beentranslated to the real-world practice of data-driven decision making. As such,there is still an urgent need to create a viable tool to facilitate fairdecision making. We propose FairSight, a visual analytic system to address thisneed; it is designed to achieve different notions of fairness in rankingdecisions through identifying the required actions -- understanding, measuring,diagnosing and mitigating biases -- that together lead to fairer decisionmaking. Through a case study and user study, we demonstrate that the proposedvisual analytic and diagnostic modules in the system are effective inunderstanding the fairness-aware decision pipeline and obtaining more fairoutcomes.

 

Quick Read (beta)

This feature is not avaialbe for this paper.