https://repositorio.ufms.br/handle/123456789/7949 2026-03-30T16:17:35Z https://repositorio.ufms.br/handle/123456789/13279 Título: ABORDAGEM DINÂMICA DE PRODUÇÃO DE CONTEÚDO LITERÁRIO UTILIZANDO IAS DE CONTROLE Abstract: This undergraduate thesis investigates the capacity of generative artificial intelligence (AI) in the production and evaluation of literary content, focusing on the comparison between human and computational perceptions of quality. The study conducts a comparative analysis between an original short story, authored by the researcher, and three narratives generated by state-of-the-art models: Claude 4 Sonnet (Anthropic), Grok 4 (X), and GPT-5 (OpenAI). The methodology began with the creation of an original short story with a pre-defined theme and Three-Act Structure. Subsequently, prompt engineering was applied to instruct the AI models to generate analogous texts in a single attempt. The resulting works were evaluated by a panel of non-expert human judges and by other AI models, based on criteria of novelty, surprise, diversity, and linguistic complexity. This study aims to contribute to the debate regarding the potentials and challenges of AI in cultural production, offering insights into the creative collaboration between humans and machines. Tipo: Trabalho de Conclusão de Curso 2025-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/13238 Título: Análise e comparação de técnicas de processamento de linguagem natural aplicadas à tradução automática de texto Abstract: This work presents a comparative analysis between two prominent Neu ral Machine Translation (NMT) architectures: Recurrent Neural Networks (RNN) with GRU units and the Transformer architecture. The study starts from the hy pothesis that Transformers, due to their use of attention mechanisms, outperform recurrent approaches in terms of performance and generalization capacity. To test this hypothesis, two prototypes were implemented: an RNN-GRU model trained from scratch for English-to-French translation, which achieved 93.89% validation accuracy but showed limitations with unknown vocabulary and morphological com plexity; and a pre-trained Transformer model (mBART-50) for the same task. The results confirmed the superiority of the Transformer, which achieved higher per formance in metrics such as BLEU (0.8687) and METEOR (0.9329), generating fluent and accurate translations. The analysis concludes that the Transformer ar chitecture—widely recognized in the literature as the state of the art in machine translation (VASWANI et al., 2017)—offers a more robust and efficient solution for capturing the syntactic and semantic nuances of languages. Tipo: Trabalho de Conclusão de Curso 2025-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/12348 Título: Sistema de Aforamentos Abstract: Desenvolvimento de um sistema em Java para gerenciar aforamentos do cemitério municipal de Ponta Porã, visando substituir processos manuais e melhorar o controle, consulta e emissão de documentos. Tipo: Trabalho de Conclusão de Curso 2025-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/12347 Título: Design de Interface Adaptável para Crianças com Transtorno do Espectro Autista Abstract: This study investigates interface design guidelines for children with Autism Spectrum Disorder (ASD), emphasizing personalization as a key accessibility resource. The research was conducted through a literature review and the development of visual prototypes based on specific design recommendations. Static screens were created to illustrate different customization criteria, such as font type and size, color schemes, and sensory stimuli. The study reinforces the importance of adaptable solutions to address the diversity of profiles within the spectrum, highlighting personalization as a central element in the development of more inclusive and effective interfaces. Tipo: Trabalho de Conclusão de Curso 2025-01-01T00:00:00Z