https://repositorio.ufu.br/handle/123456789/17904 2026-08-09T09:01:10Z 2026-08-09T09:01:10Z https://repositorio.ufu.br/handle/123456789/49371 2026-08-08T06:23:14Z 2026-03-25T00:00:00Z

Title: Uma abordagem interpretável para a detecção de ransomware baseada no consenso de SHAP Abstract: This study investigates the detection of ransomware malware based on artifacts extracted from memory using machine learning techniques. The Canadian Institute for Cybersecurity Malware Memory Dataset 2022 (CIC-MalMem-2022) was used, containing behavioral information on benign processes and ransomware variants. Initially, data preprocessing was performed, including class balancing, the removal of irrelevant attributes, and the elimination of redundancies among highly correlated features. Subsequently, four supervised algorithms—Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and AdaBoost—were evaluated through 10-fold stratified cross-validation, using metrics such as accuracy, precision, recall, F1-score, and confusion matrices. Subsequently, the explainability technique SHapley Additive exPlanations (SHAP) was applied to identify the importance of the features used by the models. Based on this analysis, a reduced set of attributes was generated through consensus among the models, enabling retraining and performance comparison. The results demonstrated that the reduced feature set maintained performance equivalent to that of the complete set, achieving an accuracy above 99.9% across all evaluated models. It is concluded that explainability-based feature selection can reduce data dimensionality without compromising detection capability, favoring lighter and more interpretable solutions for ransomware detection.

2026-03-25T00:00:00Z https://repositorio.ufu.br/handle/123456789/48841 2026-07-15T06:26:02Z 2026-03-09T00:00:00Z

Title: Extração de tópicos de transcrições de vídeos do YouTube sobre desenvolvimento de software com LLMs

2026-03-09T00:00:00Z https://repositorio.ufu.br/handle/123456789/48814 2026-07-15T06:26:08Z 2026-03-25T00:00:00Z

Title: RISCraft-V: uso do Minecraft para a materialização de uma experiência que suporte o ensino de sistemas digitais e arquitetura e organização de computadores Abstract: The reinvention of learning processes has been a constant subject of investigation in contemporary society, especially given the current model of social networks, which is based on immediate reward stimuli, imposing the need to incorporate methodologies capable of making learning more interactive and fostering active student engagement. In this context, approaches aimed at redesigning educational processes with a focus on interactivity, such as serious games, have gained ground in the field of informatics in education, especially in teaching subjects that involve low-level computing, such as Digital Systems and Computer Architecture and Organization, which, by their very nature, require a greater capacity for abstraction on the part of students. The use of serious games proves to be an effective step towards the reinvention of learning processes, becoming even more productive when clearly supported by learning theories and by the use of narratives that favor immersion. However, the commonly adopted serious game experiences, for the most part, lack the combination of an explicit theoretical foundation in learning theories and the application of a non-disruptive learning assessment method, such as Silent Assessment. Given this context, this work aims to develop an immersive experience supported by the theory of socio-interactionism, which addresses fundamental concepts from the syllabus of Digital Systems (DS) and Computer Architecture and Organization (CAO) courses using Minecraft. RISCraft-V, when compared to other similar works, uniquely incorporates socio-interactionism, narrative-building tools, a multidisciplinary approach, serious games, and Silent Assessment for the development of an educational experience. The obtained results point to the potential of the experience to enrich the learning process in DS and CAO courses, also highlighting the possibility of adapting the experience towards an even more comprehensive approach.

2026-03-25T00:00:00Z https://repositorio.ufu.br/handle/123456789/48631 2026-04-11T06:31:31Z 2026-03-20T00:00:00Z

Title: Visualização e Interação no Tesouro Direto: Uma Análise de Usabilidade e Acessibilidade na Seção de Histórico de Preços e Taxas

2026-03-20T00:00:00Z