https://repositorio.ufu.br/handle/123456789/5142 2026-10-08T18:57:30Z 2026-10-08T18:57:30Z https://repositorio.ufu.br/handle/123456789/50564 2026-10-08T12:15:29Z 2026-08-10T00:00:00Z

Title: Arquitetura reconfigurável (FPGA) baseada em modelo de Machine Learning para detecção de Ransomware em ambientes pervasivos Abstract: The growing number of ransomware attacks has put data security and service availabi lity at risk worldwide, causing an estimated financial impact of billions of dollars annually. Traditional signature-based detection approaches struggle to identify new variants due to obfuscation and polymorphism, limiting their effectiveness against zero-day attacks. At the same time, general-purpose processor architectures lack the hardware-level reconfi guration flexibility required for adaptive processing of Machine Learning (ML) models in pervasive computing environments, which are characterized by devices with limited processing, memory, and energy resources. This work proposes a ransomware detection architecture based on Field-Programmable Gate Array (FPGA) that combines Machine Learning techniques with the implementation of a classifier as combinational logic, priori tizing deterministic, low-latency inference for resource-constrained devices. The proposed method comprises training and comparing five supervised models (Decision Tree, Random Forest, eXtreme Gradient Boosting (XGBoost), Neural Network and Support Vector Ma chine (SVM)) on the CIC-MalMem-2022 dataset, using an 80/20 hold-out split combined with 5-fold cross-validation, followed by an explainability analysis using SHAP (SHa pley Additive exPlanations) to identify the consensus among models regarding the most relevant features for detection. The model that best balances predictive performance and structural simplicity will then be translated into a VHSIC Hardware Description Language (VHDL) description, in which the learned decision thresholds are embedded as constants in fixed-point comparators, enabling the synthesis of a single-cycle FPGA classifier. This work is expected to contribute to the state of the art in ransomware de tection by proposing a lightweight, low-latency hardware model suitable for embedded and pervasive devices with constrained computational resources, as an alternative to the high-performance, data-center-oriented architectures that predominate in the literature.

2026-08-10T00:00:00Z https://repositorio.ufu.br/handle/123456789/50544 2026-10-07T06:21:05Z 2026-09-08T00:00:00Z

Title: Módulo de avaliação semiautomática do sistema de candidatura em processos seletivos da pós-graduação Abstract: Graduate program admission processes involve critical stages of application submission and documentary and curricular evaluation, the latter of which is frequently carried out manually, potentially leading to inconsistencies, human error, low traceability and high operational effort. In the context of the Universidade Federal de Uberlândia (UFU), the SIPPGCO-FACOM system automated the application submission stage; however, the evaluation phase still presented limitations related to standardization and efficiency. This work presents the development of a semi-automatic evaluation module integrated into the existing system, aimed at supporting the evaluation committee in performing repetitive and systematic tasks, leaving to the evaluators only those that depend on judgement of merit. The solution encompasses an initial document screening stage, in which every required document is individually reviewed and each rejection is justified; the automatic calculation of curricular scores based on the rules defined in the selection notice; evaluation stages with no associated document, such as the interview, whose grades are recorded by the committee and combined through a weighted sum; manual adjustments with immediate score reprocessing; the generation of the candidate ranking, which combines the cut-off score and the number of positions available in each research line; and the production of structured reports, thereby promoting greater transparency, consistency, and auditability of the process. The adopted architecture follows a layered approach, composed of an Angular web application and a REST API built with TypeScript and NestJS, persisting data in a PostgreSQL database shared with the submission system. The system addresses functional and non-functional requirements related to usability, performance, security, and reliability — including the segregation between the coordination and evaluation committee profiles, a complete audit trail and data protection compliance guidelines — and was validated through automated unit and integration tests, complemented by functional testing of the user interface. The solution increases decision traceability and standardizes validations, while maintaining the flexibility required for evaluation committee intervention, and its expected benefits — reduced manual effort and fewer inconsistencies — remain to be confirmed through use in real admission processes. Furthermore, this work presents a reusable and adaptable architecture applicable to other evaluation contexts in academic admission processes.

2026-09-08T00:00:00Z https://repositorio.ufu.br/handle/123456789/50540 2026-10-07T06:20:53Z 2026-08-25T00:00:00Z

Title: Internet quântica: uma análise da evolução do conhecimento científico Abstract: The Quantum Internet has been recognized as one of the key technologies under development for the next generation of communication networks, as it exploits properties of quantum mechanics—such as superposition and entanglement—to enable new models of communication, security, and distributed computing. In this context, this dissertation aims to characterize the evolution of the Quantum Internet through an analysis of scientific knowledge over time. To this end, articles indexed in the Web of Science and Scopus databases were analyzed. This dataset was used to characterize scientific knowledge in the field through descriptive statistics and to investigate the evolution of knowledge based on citation network analysis and the identification of knowledge pathways. The results showed an increase in scientific knowledge on the topic, concentrated primarily in highly influential journals and in research groups located in China, Europe, and the United States. The knowledge pathway made it possible to analyze the scientific evolution of the Quantum Internet across three distinct layers: foundational, intermediate, and emerging. A transition was observed from studies focused on quantum communication theory to research directed toward the development of architectures, protocols, experimental networks, and distributed applications. It is concluded that the Quantum Internet has established itself as a rapidly growing field, with advances demonstrating maturity in research related to the development of large-scale quantum networks.

2026-08-25T00:00:00Z https://repositorio.ufu.br/handle/123456789/50535 2026-10-06T06:20:36Z 2026-09-11T00:00:00Z

Title: Modelo de propagação da COVID-19 em ambientes fechados baseado em autômatos celulares Abstract: COVID-19 has highlighted the importance of understanding the spread of contagious diseases in confined environments, where the proximity and movement of people favor viral transmission. This work aims to develop and evaluate an epidemiological model based on a cellular automaton to simulate the spread of COVID-19 in controlled environments, taking the study by Cavalcante et al. (2021) as a reference. The reference model was understood and implemented in the C language, with the addition of a real-time visualization through the raylib library. The cellular automaton was implemented to represent an SI (Susceptible and Infected) epidemiological model based on the Moore neighborhood and on the pseudorandom movement of individuals throughout the monitored environment. The evaluation was carried out in three stages: reproduction of the experiments from the reference paper, variation of parameters, and variation of the environment layout. The results reproduced the behavior reported in the reference, confirming that both an increase in the number of individuals and in the transmission rate intensify the contagion, while reduced populations result in almost nonexistent transmission. Furthermore, keeping the population density constant, it was found that the spatial structure of the environment does not influence the final outcome of the simulation. These results show that the developed model is able to satisfactorily reproduce the transmission dynamics of COVID-19 in confined environments, reinforcing the importance of social distancing as a control measure.

2026-09-11T00:00:00Z