Artificial Intelligence and linguistic analysis
an analysis activity using DeepSeek-R1
DOI:
https://doi.org/10.14393/DLv19a2025-53Keywords:
Linguistic Analysis, Artificial Intelligence, DeepSeek-R1, Reflection, UseAbstract
This article discusses a linguistic analysis activity carried out by DeepSeek-R1, also referred to as DeepThink (R1), a tool developed by the Chinese artificial intelligence company DeepSeek and launched in 2025. DeepSeek is a Generative Artificial Intelligence chatbot, that is, a model capable of creating texts based on prompts, similar to OpenAI’s ChatGPT. Within this system, the specific tool used here—DeepSeek-R1 or DeepThink (R1)—was selected due to its ability to develop reasoning and present it to users, thereby making explicit the path taken to produce a given response. This tool is expected to stand out in complex and challenging tasks because of its “Mixture of Experts” (MoE) approach, which processes the prompt through exclusive parameters. In the command given to this chatbot, it was asked to analyze the same sentence used in the course “Linguistic Analysis in the Teaching and Learning Contexts of Portuguese Language”, taught by Prof. Dr. Sinval Martins de Sousa Filho in the Graduate Program in Letters and Linguistics at the Federal University of Goiás. The chosen prompt was strategic, as it involved analyzing a sentence without a defined context, containing ambiguities and vagueness. The aim of this study was to observe how a trained tool conceives Linguistic Analysis (LA) and how it carries out reflection procedures on language use, in comparison with the work carried out by students in the course focused on LA. In this regard, the research is grounded in the works of Bakhtin (1997), Franchi (1987), Geraldi (1984), as well as other scholars of Linguistic Analysis such as Lima and Sousa Filho (2023) and Mendonça (2006). The data reveal limitations in DeepSeek-R1’s analysis and a conception of Linguistic Analysis that diverges from that proposed by Geraldi (1984), which highlights the importance of understanding Artificial Intelligence as a tool that works in partnership with users, rather than as a provider of irrefutable answers that exempt critical examination.
Downloads
References
BAKHTIN, M. Estética da criação verbal. 2. ed. São Paulo: Martins Fontes, 1997.
BAKHTIN, M.; VOLOCHÍNOV, V. N. Marxismo e Filosofia da Linguagem: Problemas fundamentais do método sociológico na ciência da linguagem. Tradução de Michel Lahud e Yara F. Vieira. 12. ed. São Paulo: Hucitec, 2006 [1929].
BOA SORTE, P.; FARIAS, M. A. de F.; SANTOS, A. E.; SANTOS, J. C. A.; DIAS, J. S. S. R. Inteligência artificial e escrita acadêmica: o que nos reserva o algoritmo GPT-3?. Revista EntreLinguas, Araraquara, v. 7, n. 00, p. e021035, 2021. DOI https://doi.org/10.29051/el.v7i00.15352
CARVALHO, A. S. M. et al. O impacto do Chat GPT nas práticas do ensino superior. IV Seven Internacional Multidisciplinary Congress, 2023. DOI https://doi.org/10.56238/sevenIVmulti2023-010
CHEIN, C. DeepSeek: o modelo de IA chinês que superou as sanções dos EUA. MIT Technology Review Brasil, 2025. Disponível em: https://mittechreview.com.br/deepseek-ia-china/?utm_source=chatgpt.com.
DEEPSEEK: O QUE É, COMO FUNCIONA E COMO USAR?. Divia, 2025. Disponível em: https://www.divia.com.br/deepseek-o-que-e-como-funciona-e-como-usar?utm_source=chatgpt.com.
FRANCHI, C. Criatividade e gramática. Trabalhos em linguística aplicada, v. 9, n. 1, p. 5-45, 1987.
FRANCHI, C. Linguagem-atividade constitutiva. Revista do GEL, p. 37-74, 2002.
GERALDI, J. W. et al. O texto na sala de aula. Cascavel: Assoeste, 1984.
GERALDI, J. W. Portos de passagem. Martins Fontes, 1991.
GUERRA, F.; FTEAM. A relação entre IA e linguística computacional. Folha de Pernambuco, 2023. Disponível em: https://www.folhape.com.br/colunistas/tecnologia-e-games/a-relacao-entre-ia-e-linguistica-computacional/40723/.
GROSSI, M. G. R. et al. Contribuições da inteligência artificial para a educação: Uma entrevista com o chatgpt. SYNTHESIS| Revistal Digital FAPAM, v. 12, n. 1, p. 1-20, 2023. DOI https://doi.org/10.63054/2177-823X.2023.635
INTRODUCING OPENAI o1 - PREVIEW. OpenAI, 2024. Disponível em: https://openai.com/index/introducing-openai-o1-preview/.
KLEINA, O. A diferença entre machine learning e deep learning. Pós PUC-PR Digital, 2023. Disponível em: https://posdigital.pucpr.br/blog/machine-learning-deep-learning.
KRAEMER, M. A. D.; COSTA-HUBES, T. C. Práticas de Análise Linguística na BNCC: Heterogeneidade constitutiva e marcada no discuros. In: PEREIRA, R. A.; COSTA-HUBES, T. C. (org.). Prática de análise linguística nas aulas de Língua Portuguesa. São Carlos: Pedro & João Editores, 529p., 2021.
LIMA, S. A.; SOUSA FILHO, S. M. Fundamentos da Análise Linguística Enunciativa: Humboldt, Bakhtin, Franchi e a questão do estritamente linguístico. Boletim de Conjuntura (BOCA), v. 15, n. 44, 2023. p. 559-586
MENDONÇA, M. R. S. Análise linguística no ensino médio: um novo olhar, um outro objeto. In: BUNZEN, C.; MENDONÇA, M. (org.) et al. Português no ensino médio e formação do professor. São Paulo: Parábola Editorial, 2006.
MOLON, N. D.; VIANNA, R. O círculo de Bakhtin e a linguística aplicada. Bakhtiniana: revista de estudos do discurso, v. 7, p. 142-165, 2012. DOI https://doi.org/10.1590/S2176-45732012000200010
O QUE É A IA GENERATIVA? Amazon Web Services, 2024. Disponível em: https://aws.amazon.com/pt/what-is/generative-ai/#:~:text=A%20IA%20generativa%2C%20ou%20gen,sabe%20para%20resolver%20novos%20problemas.
O QUE É GRANDE MODELO DE LINGUAGEM (LLM)? Amazon Web Services, 2024. Disponível em: https://aws.amazon.com/pt/what-is/large-language-model/.
POLATO, A. D. M.; MENEGASSI, R. J. Epistemologia teórica do nascimento da prática de análise linguística: Décadas de 80 e 90. In: PEREIRA, R. A.; COSTA-HUBES, T. C. (org.). Prática de análise linguística nas aulas de Língua Portuguesa. São Carlos: Pedro & João Editores, 529 p., 2021.
TEIXEIRA, J. O que é inteligência artificial. Rio de Janeiro: Paulus, 2019.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Lavínia Eugênio Cirqueira Silva
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish in this journal agree to the following terms:
Authors retain the copyright and waiver the journal the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), allowing the sharing of work with authorship recognition and preventing its commercial use.
Authors are authorized to take additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (publish in institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
How to Cite
