Chapter

Machine Learning and Knowledge Discovery in Databases

Volume 8725 of the series Lecture Notes in Computer Science pp 694-710

Anti-discrimination Analysis Using Privacy Attack Strategies

  • Salvatore RuggieriAffiliated withUniversità di Pisa
  • , Sara HajianAffiliated withUniversitat Rovira i Virgili
  • , Faisal KamiranAffiliated withInformation Technology, University of the Punjab
  • , Xiangliang ZhangAffiliated withKing Abdullah University of Science and Technology

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Social discrimination discovery from data is an important task to identify illegal and unethical discriminatory patterns towards protected-by-law groups, e.g., ethnic minorities. We deploy privacy attack strategies as tools for discrimination discovery under hard assumptions which have rarely tackled in the literature: indirect discrimination discovery, privacy-aware discrimination discovery, and discrimination data recovery. The intuition comes from the intriguing parallel between the role of the anti-discrimination authority in the three scenarios above and the role of an attacker in private data publishing. We design strategies and algorithms inspired/based on Frèchet bounds attacks, attribute inference attacks, and minimality attacks to the purpose of unveiling hidden discriminatory practices. Experimental results show that they can be effective tools in the hands of anti-discrimination authorities.