Gender bias in machine translation: an analysis of Google Translate in English and Spanish

  1. Lopez Medel, Maria 1
  1. 1 Universitat d'Alacant
    info

    Universitat d'Alacant

    Alicante, España

    ROR https://ror.org/05t8bcz72

Revista:
Academia Letters

ISSN: 2771-9359

Año de publicación: 2021

Tipo: Artículo

DOI: 10.20935/AL2288 GOOGLE SCHOLAR lock_openAcceso abierto editor

Otras publicaciones en: Academia Letters

Resumen

Although Google has attempted on several occasions to remove gender bias from its free online translation service, it still tends to exhibit predominantly masculine options and shows a tendency towards perpetuating or exaggerating sexist stereotypes, which adds to other flaws like a failure to notice text formality, typos and nuances. We analyse the gender bias of Google Translate between English and Spanish by entering a number of gender-invariable Spanish nouns whose referent’s gender was unknown due to the omission of pronouns and other particles. The results show a strong gender bias which needs to be removed by training machine translation software to infer semantic gender from pronouns and word terminations, and applying gender-tagging in multilingual corpora.

Referencias bibliográficas

  • Bond, Esther. (2020) “Google fixes gender bias in Google Translate (again)”. Slator. https://slator.com/machine-translation/google-fixes-gender-bias-in-google-translate-again/
  • Crespo Miguel, Mario & Marta Sánchez-Saus Laserna. (2016) “Graded acceptance in corpusbased English-to-Spanish machine translation evaluation. In: CILC 2016. EPiC Series in language and linguistics, pp. 58-70.
  • European Institute for Gender Equality. (2019) “Using different adjectives for women and men”. In: EIGE. 2019. Toolkit on gender-sensitive communication. Luxembourg: Publications Office of the European Union, p. 24. https://eige.europa.eu/publications/gendersensitivecommunication/challenges/stereotypes/using-different-adjectives-women-and-men
  • Google. (2021) Google Translate help. https://support.google.com/translate#topic=7011755
  • Guillou et. al. (2016). “Shared task: Cross-lingual pronoun prediction”. ACL 2016 First conference on machine translation (WMT16). http://statmt.org/wmt16/pronoun-task.html
  • Johnson, Melvin. (2018) “Providing gender-specific translations in Google Translate”. Google AI blog. https://ai.googleblog.com/2018/12/providing-gender-specific-translations.html
  • Johnson, Melvin. (2020) “A scalable approach to reducing gender bias in Google Translate”. Google AI blog. https://ai.googleblog.com/2020/04/a-scalable-approach-to-reducing-gender.html
  • Kayser-Bril, Nicolas. (2020) “Female historians and male nurses do not exist, Google Translate tells its European users”. Algorithm Watch. https://algorithmwatch.org/en/story/googletranslategender-bias/
  • Kuczmarski, James. (2018) “Reducing gender bias in Google Translate”. Google blog. https://blog.google/products/translate/reducing-gender-bias-google-translate/
  • Monti, Johanna. (2020) “Gender issues in machine translation. An unsolved problem?”. In: Von Flotow, Luise & Hala Kamal (eds.) 2020. The Routledge handbook of translation, feminism and gender. London: Routledge, pp. 457-468.
  • O. R. Prates, Marcelo; Pedro H. Avelar & Luís C. Lamb. (2020) “Assessing gender bias in machine translation: a case study with Google Translate”. Neural computing and applications 32, pp. 6.363-6.381.
  • Schiebinger, Londa. (2014). “Machine translation: analyzing gender”. Gendered innovations. http://genderedinnovations.stanford.edu/case-studies/nlp.html#tabs-1
  • Weaver, Warren. (1949) “Translation”. Machine translation of languages 14, pp. 15-23. http://www.mt-archive.info/Weaver-1949.pdf