A call for a research agenda on fair NLP for Portuguese
Resumo
Diverse areas widely apply artificial intelligence and natural language processing (NLP) tools to their contexts. However, these algorithms present ethical issues, such as biased and discriminatory decisions. For example, representation biases in NLP can result in discriminatory behavior towards race and gender. Works have been addressing this issue and seeking to build fair NLP solutions, however they mainly focus on Anglo-Saxon languages. This work aims to challenge the scientific community in order to stimulate and motivate further research in the fair NLP specifically for the Portuguese language. To achieve this, a literature review was conducted to identify existing research efforts and indicate future directions.
Palavras-chave:
Fairness, Natural Language Processing, Portuguese
Referências
Araujo, R., Fornazin, M., and Pimentel, M. (2017). Uma análise sobre a produção de conhecimento científico nas pesquisas publicadas nos primeiros 10 anos da isys (2008-2017). iSys-Brazilian Journal of Information Systems, 10(4):45–65. https://doi.org/10.5753/isys.2017.351
Bender, E. M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? 🦜. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, FAccT ’21, page 610–623, New York, NY, USA. Association for Computing Machinery. https://doi.org/10.1145/3442188.3445922
Blodgett, S. L., Barocas, S., Daumé III, H., and Wallach, H. (2020). Language (technology) is power: A critical survey of” bias” in nlp. arXiv preprint arXiv:2005.14050.
Blodgett, S. L. and O’Connor, B. (2017). Racial disparity in natural language processing: A case study of social media african-american english. arXiv preprint arXiv:1707.00061.
Bolukbasi, T., Chang, K.-W., Zou, J. Y., Saligrama, V., and Kalai, A. T. (2016). Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in neural information processing systems, 29.
Buolamwini, J. and Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. In Conference on fairness, accountability and transparency, pages 77–91. PMLR.
Camões - Instituto da Cooperação e da Língua (2023). Dados sobre a língua portuguesa. [link].
Cho, W. I., Kim, J., Yang, J., and Kim, N. S. (2021). Towards cross-lingual generalization of translation gender bias. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pages 449–457. https://doi.org/10.1145/3442188.3445907
Dastin, J. (2018). Amazon scraps secret ai recruiting tool that showed bias against women. Reuters. [link]
Drager, K., Rilliard, A. O. B., Vieira, M. d. S. M., and Wiedemer, M. L. (2021). Linguistic varieties in brazil and beyond. Revista Diadorim, 23(1):24–33. https://doi.org/10.35520/diadorim.2021.v23n1a44441
Font, J. E. and Costa-Jussa, M. R. (2019). Equalizing gender biases in neural machine translation with word embeddings techniques. arXiv preprint arXiv:1901.03116.
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Iii, H. D., and Crawford, K. (2021). Datasheets for datasets. Communications of the ACM, 64(12):86–92. https://doi.org/10.1145/3458723
Guy, G. R. (1981). Linguistic variation in brazilian portuguese: Aspects of the phonology, syntax, and language history.
Hartmann, N., Fonseca, E., Shulby, C., Treviso, M., Rodrigues, J., and Aluisio, S. (2017). Portuguese word embeddings: Evaluating on word analogies and natural language tasks. arXiv preprint arXiv:1708.06025.
Leavy, S. (2018). Gender bias in artificial intelligence: The need for diversity and gender theory in machine learning. In Proceedings of the 1st international workshop on gender equality in software engineering, pages 14–16. https://doi.org/10.1145/3195570.3195580
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., and Galstyan, A. (2021). A survey on bias and fairness in machine learning. ACM Computing Surveys (CSUR), 54(6):1–35. https://doi.org/10.1145/3457607
Ruback, L., Avila, S., and Cantero, L. (2021). Vieses no aprendizado de máquina e suas implicações sociais: Um estudo de caso no reconhecimento facial. In Anais do II Workshop sobre as Implicações da Computação na Sociedade, pages 90–101, Porto Alegre, RS, Brasil. SBC. https://doi.org/10.5753/wics.2021.15967
Santana, B. S., Woloszyn, V., and Wives, L. K. (2018). Is there gender bias and stereotype in portuguese word embeddings? arXiv preprint arXiv:1810.04528.
Silva, T. (2020). Visão computacional e racismo algorítmico: branquitude e opacidade no aprendizado de máquina. Revista ABPN, 12:428–448.
Sun, T., Gaut, A., Tang, S., Huang, Y., ElSherief, M., Zhao, J., Mirza, D., Belding, E., Chang, K.-W., and Wang, W. Y. (2019). Mitigating gender bias in natural language processing: Literature review. arXiv preprint arXiv:1906.08976.
Vanmassenhove, E., Hardmeier, C., and Way, A. (2018). Getting gender right in neural machine translation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 3003–3008, Brussels, Belgium. Association for Computational Linguistics. https://doi.org/10.18653/v1/D18-1334
Bender, E. M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? 🦜. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, FAccT ’21, page 610–623, New York, NY, USA. Association for Computing Machinery. https://doi.org/10.1145/3442188.3445922
Blodgett, S. L., Barocas, S., Daumé III, H., and Wallach, H. (2020). Language (technology) is power: A critical survey of” bias” in nlp. arXiv preprint arXiv:2005.14050.
Blodgett, S. L. and O’Connor, B. (2017). Racial disparity in natural language processing: A case study of social media african-american english. arXiv preprint arXiv:1707.00061.
Bolukbasi, T., Chang, K.-W., Zou, J. Y., Saligrama, V., and Kalai, A. T. (2016). Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in neural information processing systems, 29.
Buolamwini, J. and Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. In Conference on fairness, accountability and transparency, pages 77–91. PMLR.
Camões - Instituto da Cooperação e da Língua (2023). Dados sobre a língua portuguesa. [link].
Cho, W. I., Kim, J., Yang, J., and Kim, N. S. (2021). Towards cross-lingual generalization of translation gender bias. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pages 449–457. https://doi.org/10.1145/3442188.3445907
Dastin, J. (2018). Amazon scraps secret ai recruiting tool that showed bias against women. Reuters. [link]
Drager, K., Rilliard, A. O. B., Vieira, M. d. S. M., and Wiedemer, M. L. (2021). Linguistic varieties in brazil and beyond. Revista Diadorim, 23(1):24–33. https://doi.org/10.35520/diadorim.2021.v23n1a44441
Font, J. E. and Costa-Jussa, M. R. (2019). Equalizing gender biases in neural machine translation with word embeddings techniques. arXiv preprint arXiv:1901.03116.
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Iii, H. D., and Crawford, K. (2021). Datasheets for datasets. Communications of the ACM, 64(12):86–92. https://doi.org/10.1145/3458723
Guy, G. R. (1981). Linguistic variation in brazilian portuguese: Aspects of the phonology, syntax, and language history.
Hartmann, N., Fonseca, E., Shulby, C., Treviso, M., Rodrigues, J., and Aluisio, S. (2017). Portuguese word embeddings: Evaluating on word analogies and natural language tasks. arXiv preprint arXiv:1708.06025.
Leavy, S. (2018). Gender bias in artificial intelligence: The need for diversity and gender theory in machine learning. In Proceedings of the 1st international workshop on gender equality in software engineering, pages 14–16. https://doi.org/10.1145/3195570.3195580
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., and Galstyan, A. (2021). A survey on bias and fairness in machine learning. ACM Computing Surveys (CSUR), 54(6):1–35. https://doi.org/10.1145/3457607
Ruback, L., Avila, S., and Cantero, L. (2021). Vieses no aprendizado de máquina e suas implicações sociais: Um estudo de caso no reconhecimento facial. In Anais do II Workshop sobre as Implicações da Computação na Sociedade, pages 90–101, Porto Alegre, RS, Brasil. SBC. https://doi.org/10.5753/wics.2021.15967
Santana, B. S., Woloszyn, V., and Wives, L. K. (2018). Is there gender bias and stereotype in portuguese word embeddings? arXiv preprint arXiv:1810.04528.
Silva, T. (2020). Visão computacional e racismo algorítmico: branquitude e opacidade no aprendizado de máquina. Revista ABPN, 12:428–448.
Sun, T., Gaut, A., Tang, S., Huang, Y., ElSherief, M., Zhao, J., Mirza, D., Belding, E., Chang, K.-W., and Wang, W. Y. (2019). Mitigating gender bias in natural language processing: Literature review. arXiv preprint arXiv:1906.08976.
Vanmassenhove, E., Hardmeier, C., and Way, A. (2018). Getting gender right in neural machine translation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 3003–3008, Brussels, Belgium. Association for Computational Linguistics. https://doi.org/10.18653/v1/D18-1334
Publicado
25/09/2023
Como Citar
DE LIMA, Luiz Fernando F. P.; DE ARAUJO, Renata Mendes.
A call for a research agenda on fair NLP for Portuguese. In: SIMPÓSIO BRASILEIRO DE TECNOLOGIA DA INFORMAÇÃO E DA LINGUAGEM HUMANA (STIL), 14. , 2023, Belo Horizonte/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2023
.
p. 187-192.
DOI: https://doi.org/10.5753/stil.2023.233763.
