Detecção de Discurso de Ódio: Homofobia
Resumo
Entender o contexto e os motivos para ações homofóbicas é primordial para mitigar esse tipo de ódio. Este trabalho apresenta a construção de um conjunto de dados retirados do Twitter, os quais contém informações sobre discursos homofóbicos. As contribuições são: (1) o método de construção de um conjunto de dados e a rotulação anônima baseada no teor homofóbico dos mesmos; (2) a criação de características desse conjunto de dados; (3) a avaliação de métodos de aprendizagem de máquina para a classificação dos dados referente ao teor homofóbico. Resultados preliminares mostram que o modelo Random Forest Classifier se destaca na identificação de tuítes homofóbicos com F1-Score de 0.8, revocação de 0.9 e precisão de 0.7.
Palavras-chave:
homofobia, discurso de ódio, ciência de dados
Referências
Affde (2021). Quantas pessoas usam o twitter em 2021? [novas estatísticas do twitter]. https://www.affde.com/pt/twitter-users.html.
Amper (2021). We are social e hootsuite digital 2021 [resumo e relatório completo]. https://tinyurl.com/we-are-social-e-hootsuite.
Bermingham, A. and Smeaton, A. (2011). On using Twitter to monitor political sentiment and predict election results. In Proceedings of the Workshop on Sentiment Analysis where AI meets Psychology (SAAIP 2011), pages 2–10, Chiang Mai, Thailand. Asian Federation of Natural Language Processing.
Burnap, P. and Williams, M. L. (2015). Cyber hate speech on twitter: An application of machine classification and statistical modeling for policy and decision making. Policy & Internet, 7(2):223–242.
Cazelatto, C. and Cardin, V. S. G. (2016). O discurso de ódio homofóbico no brasil: um instrumento limitador da sexualidade humana. Revista Jurídica Cesumar-Mestrado, 16.
Chamberland and Lebreton (2012). Réflexions autour de la notion d’homophobie : succés politique, malaises conceptuels et application empirique. Nouv Questions Feministes.
Chambers, S. A. (2007). An incalculable effect’: Subversions of heteronormativity. Political studies, 55.
Council, H. R. (2013). Human rights. annual report of the united nations high commissioner for human rights. https://tinyurl.com/rabat-united-nations.
Davidson, T., Warmsley, D., Macy, M. W., and Weber, I. (2017). Automated hate speech detection and the problem of offensive language. CoRR, abs/1703.04009.
Frässé, C. and Barrientos, J. (2016). The concept of homophobia: A psychosocial perspective. Sexologies, 25.
Kwok, I. and Wang, Y. (2013). Locate the hate: Detecting tweets against blacks. In AAAI.
Miriam J., M. M. (2002). The trivium: The liberal arts of logic, grammar, and rhetoric : understanding the nature and function of language. Paul Dry Books.
Pereira-Kohatsu, J. C., Quijano-Sánchez, L., Liberatore, F., and Camacho-Collados, M. (2019). Detecting and monitoring hate speech in twitter. Sensors, 19(21).
Statista (2021). Leading countries based on number of twitter users as of october 2021. https://tinyurl.com/statista-twitter-users.
Watanabe, H., Bouazizi, M., and Ohtsuki, T. (2018). Hate speech on twitter: A pragmatic approach to collect hateful and offensive expressions and perform hate speech detection. IEEE Access, 6:13825–13835.
Weinberg, G. H. (1975). Society and the healthy homosexual / [by] George Weinberg. Gerrards Cross, Smythe.
Amper (2021). We are social e hootsuite digital 2021 [resumo e relatório completo]. https://tinyurl.com/we-are-social-e-hootsuite.
Bermingham, A. and Smeaton, A. (2011). On using Twitter to monitor political sentiment and predict election results. In Proceedings of the Workshop on Sentiment Analysis where AI meets Psychology (SAAIP 2011), pages 2–10, Chiang Mai, Thailand. Asian Federation of Natural Language Processing.
Burnap, P. and Williams, M. L. (2015). Cyber hate speech on twitter: An application of machine classification and statistical modeling for policy and decision making. Policy & Internet, 7(2):223–242.
Cazelatto, C. and Cardin, V. S. G. (2016). O discurso de ódio homofóbico no brasil: um instrumento limitador da sexualidade humana. Revista Jurídica Cesumar-Mestrado, 16.
Chamberland and Lebreton (2012). Réflexions autour de la notion d’homophobie : succés politique, malaises conceptuels et application empirique. Nouv Questions Feministes.
Chambers, S. A. (2007). An incalculable effect’: Subversions of heteronormativity. Political studies, 55.
Council, H. R. (2013). Human rights. annual report of the united nations high commissioner for human rights. https://tinyurl.com/rabat-united-nations.
Davidson, T., Warmsley, D., Macy, M. W., and Weber, I. (2017). Automated hate speech detection and the problem of offensive language. CoRR, abs/1703.04009.
Frässé, C. and Barrientos, J. (2016). The concept of homophobia: A psychosocial perspective. Sexologies, 25.
Kwok, I. and Wang, Y. (2013). Locate the hate: Detecting tweets against blacks. In AAAI.
Miriam J., M. M. (2002). The trivium: The liberal arts of logic, grammar, and rhetoric : understanding the nature and function of language. Paul Dry Books.
Pereira-Kohatsu, J. C., Quijano-Sánchez, L., Liberatore, F., and Camacho-Collados, M. (2019). Detecting and monitoring hate speech in twitter. Sensors, 19(21).
Statista (2021). Leading countries based on number of twitter users as of october 2021. https://tinyurl.com/statista-twitter-users.
Watanabe, H., Bouazizi, M., and Ohtsuki, T. (2018). Hate speech on twitter: A pragmatic approach to collect hateful and offensive expressions and perform hate speech detection. IEEE Access, 6:13825–13835.
Weinberg, G. H. (1975). Society and the healthy homosexual / [by] George Weinberg. Gerrards Cross, Smythe.
Publicado
31/07/2022
Como Citar
SOUZA, Andrey O.; NAKAMURA, Eduardo F.; NAKAMURA, Fabíola G..
Detecção de Discurso de Ódio: Homofobia. In: BRAZILIAN E-SCIENCE WORKSHOP (BRESCI), 16. , 2022, Niterói.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2022
.
p. 73-80.
ISSN 2763-8774.
DOI: https://doi.org/10.5753/bresci.2022.223222.
