https://repositorio.ufms.br/handle/123456789/6098 Thu, 02 Jul 2026 15:36:29 GMT 2026-07-02T15:36:29Z https://repositorio.ufms.br/handle/123456789/14588 Título: CamGuide: An Efficient Method Abstract: A review of weakly supervised learning in deep learning, using classical methods as comparative baselines to introduce a new method based on knowledge distillation and carefully calibrated heuristics for smaller networks. Tipo: Trabalho de Conclusão de Curso Thu, 01 Jan 2026 00:00:00 GMT https://repositorio.ufms.br/handle/123456789/14588 2026-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/14517 Título: Avaliação da Generalização de Modelos BERTimbau na Detecção de Desinformação em Português com Análise Qualitativa de Dados Abstract: The growing dissemination of information on the internet has intensified the spread of fake news, making automatic fake news detection a relevant problem. In this context, pre-trained language models have shown significant performance in natural language processing tasks, including text classification in Portuguese. This work evaluates the performance of the BERTimbau model in classifying news as true or false, using multiple datasets with distinct characteristics. To ensure a balanced evaluation, the datasets were previously balanced according to their classes. The experiments were conducted to analyze not only the model performance within the training domain, but also its generalization ability across different datasets. Additionally, a qualitative analysis step based on a large language model was incorporated to examine textual patterns, differences among datasets, and recurring aspects associated with the classifier's correct and incorrect predictions. The results indicate that, although the model achieves good performance on data similar to those used during training, there is a significant performance reduction in out-of-domain scenarios, highlighting limitations in generalization. These findings reinforce the importance of robustness evaluation in real-world applications of language models. Tipo: Trabalho de Conclusão de Curso Thu, 01 Jan 2026 00:00:00 GMT https://repositorio.ufms.br/handle/123456789/14517 2026-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/14503 Título: Dos Estudos em Laboratório a Aplicação Prática em Análise Não Invasiva de Segurança Abstract: This article integrates three workstreams developed throughout the academic activity: a simplified report on practical studies, a supplementary technical report on the case study, and the set of objective evidence collected during the analysis. The objective is to demonstrate not only which topics were studied but, more importantly, what was gained from these studies regarding a non-invasive analysis of the UFMS certificate system. The training phase involved network reconnaissance using Nmap, request interception and analysis with Burp Suite, the study of XSS and IDOR vulnerabilities, hash cracking with John the Ripper, and the review of alerts and logs within a monitoring environment. Based on this background, a low-impact evidence-gathering process regarding SICERT was structured, focusing on HTTP headers, session cookies, browser behavior, a local proof-of-concept for clickjacking, and logout verification using the evaluator's own session. The results showed that the primary outcome of the studies was the ability to transform laboratory content into a reproducible procedure for observation, classification, and technical recommendation, underpinned by real-world evidence and well-defined ethical boundaries. The most relevant commands and code snippets have been compiled at the end of the document in technical appendices supplementary to the main body of the article. Tipo: Trabalho de Conclusão de Curso Thu, 01 Jan 2026 00:00:00 GMT https://repositorio.ufms.br/handle/123456789/14503 2026-01-01T00:00:00Z https://repositorio.ufms.br/handle/123456789/14501 Título: Uma proposta de Data Linter para Qualidade de Dados Abstract: Uma proposta de Data Linter para Qualidade de Dados Tipo: Trabalho de Conclusão de Curso Thu, 01 Jan 2026 00:00:00 GMT https://repositorio.ufms.br/handle/123456789/14501 2026-01-01T00:00:00Z