FairRankVis: A Visual Analytics Framework for Exploring Algorithmic Fairness in Graph Mining Models

Tiankai Xie, Yuxin Ma, Jian Kang, Hanghang Tong, Ross Maciejewski

Research output: Contribution to journalArticlepeer-review

Abstract

Graph mining is an essential component of recommender systems and search engines. Outputs of graph mining models typically provide a ranked list sorted by each item's relevance or utility. However, recent research has identified issues of algorithmic bias in such models, and new graph mining algorithms have been proposed to correct for bias. As such, algorithm developers need tools that can help them uncover potential biases in their models while also exploring the impacts of correcting for biases when employing fairness-aware algorithms. In this paper, we present FairRankVis, a visual analytics framework designed to enable the exploration of multi-class bias in graph mining algorithms. We support both group and individual fairness levels of comparison. Our framework is designed to enable model developers to compare multi-class fairness between algorithms (for example, comparing PageRank with a debiased PageRank algorithm) to assess the impacts of algorithmic debiasing with respect to group and individual fairness. We demonstrate our framework through two usage scenarios inspecting algorithmic fairness.

Original languageEnglish (US)
JournalIEEE Transactions on Visualization and Computer Graphics
DOIs
StateAccepted/In press - 2021

Keywords

  • Analytical models
  • Data mining
  • Data models
  • fairness
  • Graph ranking
  • Machine learning
  • Machine learning algorithms
  • Task analysis
  • Visual analytics
  • visual analytics

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design

Fingerprint

Dive into the research topics of 'FairRankVis: A Visual Analytics Framework for Exploring Algorithmic Fairness in Graph Mining Models'. Together they form a unique fingerprint.

Cite this