Fairness in Proprietary Image Tagging Algorithms: A Cross-Platform Audit on People Images

  • Kyriakos Kyriakou Research Centre on Interactive Media, Smart Systems and Emerging Technologies
  • Pınar Barlas Research Centre on Interactive Media, Smart Systems and Emerging Technologies
  • Styliani Kleanthous Research Centre on Interactive Media, Smart Systems and Emerging Technologies
  • Jahna Otterbacher Research Centre on Interactive Media, Smart Systems and Emerging Technologies

Abstract

There are increasing expectations that algorithms should behave in a manner that is socially just. We consider the case of image tagging APIs and their interpretations of people images. Image taggers have become indispensable in our information ecosystem, facilitating new modes of visual communication and sharing. Recently, they have become widely available as Cognitive Services. But while tagging APIs offer developers an inexpensive and convenient means to add functionality to their creations, most are opaque and proprietary. Through a cross-platform comparison of six taggers, we show that behaviors differ significantly. While some offer more interpretation on images, they may exhibit less fairness toward the depicted persons, by misuse of gender-related tags and/or making judgments on a person’s physical appearance. We also discuss the difficulties of studying fairness in situations where algorithmic systems cannot be benchmarked against a ground truth.

Published
2019-07-06
How to Cite
Kyriakou, K., Barlas, P., Kleanthous, S., & Otterbacher, J. (2019). Fairness in Proprietary Image Tagging Algorithms: A Cross-Platform Audit on People Images. Proceedings of the International AAAI Conference on Web and Social Media, 13(01), 313-322. Retrieved from https://aaai.org/ojs/index.php/ICWSM/article/view/3232