Diagnosing Gender Bias in Image Recognition Systems

to connected objects

Abstract: Image recognition systems offer the promise to learn from images at scale without requiring expert knowledge. However, past research suggests that machine learning systems often produce biased output. In this article, we evaluate potential gender biases of commercial image recognition platforms using photographs of U.S. members of Congress and a large number of Twitter images posted by these politicians. Our crowdsourced validation shows that commercial image recognition systems can produce labels that are correct and biased at the same time as they selectively report a subset of many possible true labels. We find that images of women received three times more annotations related to physical appearance. Moreover, women in images are recognized at substantially lower rates in comparison with men. We discuss how encoded biases such as these affect the visibility of women, reinforce harmful gender stereotypes, and limit the validity of the insights that can be gathered from such data

Location
Deutsche Nationalbibliothek Frankfurt am Main
Extent
Online-Ressource
Language
Englisch
Notes
Veröffentlichungsversion
begutachtet (peer reviewed)
In: Socius: Sociological Research for a Dynamic World ; 6 (2020) ; 1-17

Event
Veröffentlichung
(where)
Mannheim
(who)
SSOAR, GESIS – Leibniz-Institut für Sozialwissenschaften e.V.
(when)
2020
Creator
Schwemmer, Carsten
Knight, Carly
Bello-Pardo, Emily D.
Oklobdzija, Stan
Schoonvelde, Martijn
Lockhart, Jeffrey W.

DOI
10.1177/2378023120967171
URN
urn:nbn:de:101:1-2022040508560554642198
Rights
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
Last update
15.08.2025, 7:21 AM CEST

Data provider

This object is provided by:
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Show original at data provider

Associated

  • Schwemmer, Carsten
  • Knight, Carly
  • Bello-Pardo, Emily D.
  • Oklobdzija, Stan
  • Schoonvelde, Martijn
  • Lockhart, Jeffrey W.
  • SSOAR, GESIS – Leibniz-Institut für Sozialwissenschaften e.V.

Time of origin

  • 2020

Other Objects (12)

Facial trustworthiness dampens own-gender bias in emotion recognition

Facial trustworthiness dampens own-gender bias in emotion recognition

Representativeness and face-ism : Gender bias in image search

Representativeness and face-ism : Gender bias in image search

Measuring, visualizing, and diagnosing reference bias with biastools

Measuring, visualizing, and diagnosing reference bias with biastools

Bias: a normal operating characteristic of the diagnosing brain

Bias: a normal operating characteristic of the diagnosing brain

Bias against research on gender bias

Bias against research on gender bias

Diagnosing discrimination: stock returns and CEO gender

Arbeitspapier

Diagnosing discrimination: stock returns and CEO gender

Less can be more: representational vs. stereotypical gender bias in facial expression recognition

Less can be more: representational vs. stereotypical gender bias in facial expression recognition

Translation analysis of English address image recognition based on image recognition

Translation analysis of English address image recognition based on image recognition

Online Teaching and Gender Bias

Arbeitspapier

Online Teaching and Gender Bias

Gender Bias in Teaching Evaluations

Arbeitspapier

Gender Bias in Teaching Evaluations

Gender Bias in Teaching Evaluations

Arbeitspapier

Gender Bias in Teaching Evaluations

Gender bias and credit access

Arbeitspapier

Gender bias and credit access

Related objects