A Computational Framework for Measuring and Analyzing Gender Bias in Portuguese-language Literary Texts

• Mariana O. Silva Universidade Federal de Minas Gerais (UFMG)
• Michele A. Brandão Universidade Federal de Minas Gerais (UFMG)
• Mirella M. Moro Universidade Federal de Minas Gerais (UFMG)

DOI: https://doi.org/10.5753/sbbd_estendido.2026.249414

Resumo

Literary texts often reproduce and reshape social biases, including gender bias, through recurring linguistic patterns that influence how characters are represented. While traditional literary criticism offers detailed interpretations, it is limited in scale, and most computational studies of gender bias focus on English, leaving Portuguese-language literature underexplored. This dissertation addresses this gap by proposing PORTALIA, a modular computational framework for measuring and analyzing gender bias in Portuguese-language literary texts. The framework integrates Natural Language Processing techniques for character identification, automated gender inference, linguistic analysis, and quantitative bias measurement. It is evaluated on Portuguese literary corpora and supports large-scale, reproducible analyses of gendered patterns in character representation, descriptors, and narrative agency.

Palavras-chave: NER, Portuguese-language literature, social biases, NLP

Referências

Caliskan, A., Bryson, J. J., and Narayanan, A. (2017). Semantics Derived Automatically from Language Corpora Contain Human-like Biases. Science, 356(6334):183–186.

Caton, S. and Haas, C. (2024). Fairness in Machine Learning: A Survey. ACM Comput. Surv., 56(7).

Freitas, C. and Martins, F. (2023). Bela, Recatada e do Lar: O Que a Mineração de Textos Literários Nos Diz Sobre a Caracterização de Personagens Femininas e Masculinas. Fórum Linguístico, 20(3):9118–9138.

Freitas, C. and Santos, D. (2023). Gender Depiction in Portuguese. In Conference Reader: 2nd Annual Conference of Computational Literary Studies (CCLS2023), pages 4–30.

Hoyle, A. M., Wolf-Sonkin, L., Wallach, H. M., Augenstein, I., et al. (2019). Unsupervised Discovery of Gendered Language through Latent-Variable Modeling. In Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL, pages 1706–1716. Association for Computational Linguistics.

Lima, L. F. F. P. d. and Araujo, R. M. d. (2023). A Call for a Research Agenda on Fair NLP for Portuguese. In STIL, pages 187–192. SBC.

Luo, K., Mao, Y., Zhang, B., and Hao, S. (2024). Reflecting the Male Gaze: Quantifying Female Objectification in 19th and 20th Century Novels. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 13803–13812, Torino, Italia. ELRA and ICCL.

O. Silva, M. and Moro, M. (2025). Aprimorando o Reconhecimento de Entidades Nomeadas em Textos Literários em Português com Modelos Adaptativos. Linguamática, 17(1):77–94.

Schulz, D. and Štěpán Bahník (2019). Gender Associations in the Twentieth-century English-language Literature. Journal of Research in Personality, 81:88–97.

Scofield, C., Silva, M., and Moro, M. (2022a). What makes a book successful? a study on portuguese-language literature. In Anais Estendidos do XXVIII Simpósio Brasileiro de Sistemas Multimídia e Web, pages 69–72, Porto Alegre, RS, Brasil. SBC.

Scofield, C., Silva, M. O., de Melo-Gomes, L., and Moro, M. M. (2022b). Book Genre Classification Based on Reviews of Portuguese-Language Literature. In 15th International Conference Computational Processing of the Portuguese Language - PROPOR, pages 188–197. Springer.

Silva, M., ao, M. B., and Moro, M. (2025a). Gender Bias in Portuguese Literary Texts: A Masked Language Model Approach. In Anais do XVI Simpósio Brasileiro de Tecnologia da Informação e da Linguagem Humana, pages 407–419, Porto Alegre, RS, Brasil. SBC.

Silva, M., Brandão, M., and M. Moro, M. (2025b). Gender Bias in Portuguese Literary Texts: A Masked Language Model Approach.

Silva, M., Brandão, M., and M. Moro, M. (2025c). Rewriting Stories with LLMs: Gender Bias in Generated Portuguese-language Narratives.

Silva, M. and Moro, M. (2024a). NLP Pipeline for Gender Bias Detection in Portuguese Literature. In Anais do LI Seminário Integrado de Software e Hardware (SEMISH), pages 169–180. SBC.

Silva, M. O., ao, M. A. B., and Moro, M. M. (2024a). Gender Bias Detection in Literary Texts: A Systematic Literature Review. ACM Computing Surveys. Under submission.

Silva, M. O., Brandão, M. A., and Moro, M. M. (2025d). Rewriting Stories with LLMs: Gender Bias in Generated Portuguese-language Narratives. Journal of the Brazilian Computer Society, 31(1):1120–1136.

Silva, M. O., de Melo-Gomes, L., and Moro, M. M. (2024b). From Words to Gender: Quantitative Analysis of Body Part Descriptions within Literature in Portuguese. Information Processing & Management, 61(3):103647.

Silva, M. O., Melo-Gomes, L., and Moro, M. (2023a). Gender Representation in Literature: Analysis of Characters’ Physical Descriptions. In Anais do XI Symposium on Knowledge Discovery, Mining and Learning, pages 17–24, Porto Alegre, RS, Brasil. SBC.

Silva, M. O. and Moro, M. M. (2024b). Evaluating Pre-training Strategies for Literary Named Entity Recognition in Portuguese. In Proceedings of the 16th International Conference on Computational Processing of Portuguese, PROPOR 2024, Santiago de Compostela, Galicia/Spain, 12-15 March, 2024, pages 384–393. Association for Computational Lingustics.

Silva, M. O. and Moro, M. M. (2024c). PPORTAL_ner: An Annotated Corpus of Portuguese Literary Entities. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2024, pages 12927–12937. European Language Resources Association (ELRA).

Silva, M. O. and Moro, M. M. (2024d). PPORTAL_ner: An Annotated Corpus of Portuguese Literary Entities.

Silva, M. O., Oliveira, G., and Moro, M. (2023b). Analyzing Character Networks in Portuguese-language Literary Works. In Anais do XII Brazilian Workshop on Social Network Analysis and Mining, pages 115–126, Porto Alegre, RS, Brasil. SBC.

Silva, M. O., Oliveira, G., and Moro, M. (2024c). Data Insights on Gender Representation: Analyzing the Book and Music Industries. In Anais Estendidos do XXXIX Simpósio Brasileiro de Bancos de Dados, pages 338–347. SBC.

Silva, M. O., Scofield, C., de Melo-Gomes, L., Botelho, J. E., et al. (2022a). Brazilian Reading Preferences in Goodreads: Cross-state and Cross-region Analyses. iSys - Brazilian Journal of Information Systems, 15(1):25:1–25:20.

Silva, M. O., Scofield, C., de Melo-Gomes, L., and Moro, M. M. (2022b). Cross-collection Dataset of Public Domain Portuguese-language Works. Journal of Information and Data Management, 13(1).

Silva, M. O., Scofield, C., and Moro, M. M. (2021a). PPORTAL: Public Domain Portuguese-language Literature Dataset. In Anais do III Dataset Showcase Workshop, pages 77–88, Rio de Janeiro, Brazil. SBC.

Silva, M. O., Scofield, C., Oliveira, G., Seufitelli, D., et al. (2021b). Exploring Brazilian Cultural Identity Through Reading Preferences. In Anais do X Brazilian Workshop on Social Network Analysis and Mining, pages 115–126, Porto Alegre, RS, Brasil. SBC.

Stuhler, O. (2024). The Gender Agency Gap in Fiction Writing (1850 to 2010). Proceedings of the National Academy of Sciences, 121(29):e2319514121.