Orivaldo Vieira de Santana Júnior
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Orivaldo Vieira de Santana Júnior has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Efrain Pantaleon-Matamoros
Universidade Federal do Rio Grande do Norte, Brasil
4 shared publications - Horácio Betcel Guimarães3 shared publications
- Luiz Marcos Garcia Gonçalves
Federal University of Rio Grande do Norte
3 shared publications - Aedê Cadaxa
Ministry of Health, Brasília 70070-600, Brazil
2 shared publications - Agnaldo Souza Cruz
Laboratory for Technological Innovation in Health (LAIS), Federal University of Rio Grande do Norte, Natal 59010-090, Brazil
2 shared publications - Anderson Costa Lima
Universidade Federal do Rio Grande do Norte, Brasil
2 shared publications
Latest Publications
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 98741-98751
Power Transformer Fault Classification Using Industry 4.0-Compliant Machine Learning Under Data Missing Conditions
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 98741-98751
Open AccessPower Transformer Fault Classification Using Industry 4.0-Compliant Machine Learning Under Data Missing Conditions
- Open Access
- 15 December 2025
- Published by Universidade Regional Integrada do Alto Uruguai e das Missoes in Revista ENCITEC
- Vol. 16 (1) , 7-29
USO APLICADO DE MACHINE LEARNING COM RANDOM FOREST E SELF-ORGANIZING MAP COMO MODELO PREDITIVO PARA EVASÃO DE ALUNOS
Este artigo é um recorte de uma dissertação de mestrado desenvolvida no Programa de Pós-Graduação em Ciência, Tecnologia e Inovação da Universidade Federal do Rio Grande do Norte (UFRN). Com o objetivo de investigar fatores relacionados à evasão no ensino superior por meio de técnicas de Mineração de Dados Educacionais (MDE) e Machine Learning (ML). O estudo busca explorar os dados do curso Interdisciplinar em Ciências e Tecnologia (C&T) da própria universidade, considerando os alunos ingressantes entre os anos de 2014 a 2023, com o objetivo de desenvolver modelos analíticos que identifiquem características de intervenção para o desenvolvimento acadêmico dos estudantes. O estudo combina algoritmos de ML, como Random Forest (classificação) e Self-Organizing Maps (clustering), em uma abordagem híbrida, aplicada à MDE. O resultado é um modelo preditivo capaz de identificar atributos que impactam a evasão, utilizando a explicabilidade da técnica SHapley Additive exPlanations para encontrar um conjunto de características explicáveis e ao mesmo tempo, fornecer uma melhoria significativa no poder preditivo do modelo.
- 15 December 2025
- Published by Universidade Regional Integrada do Alto Uruguai e das Missoes in Revista ENCITEC
- Vol. 16 (1) , 7-29
Open AccessUSO APLICADO DE MACHINE LEARNING COM RANDOM FOREST E SELF-ORGANIZING MAP COMO MODELO PREDITIVO PARA EVASÃO DE ALUNOS
Este artigo é um recorte de uma dissertação de mestrado desenvolvida no Programa de Pós-Graduação em Ciência, Tecnologia e Inovação da Universidade Federal do Rio Grande do Norte (UFRN). Com o objetivo de investigar fatores relacionados à evasão no ensino superior por meio de técnicas de Mineração de Dados Educacionais (MDE) e Machine Learning (ML). O estudo busca explorar os dados do curso Interdisciplinar em Ciências e Tecnologia (C&T) da própria universidade, considerando os alunos ingressantes entre os anos de 2014 a 2023, com o objetivo de desenvolver modelos analíticos que identifiquem características de intervenção para o desenvolvimento acadêmico dos estudantes. O estudo combina algoritmos de ML, como Random Forest (classificação) e Self-Organizing Maps (clustering), em uma abordagem híbrida, aplicada à MDE. O resultado é um modelo preditivo capaz de identificar atributos que impactam a evasão, utilizando a explicabilidade da técnica SHapley Additive exPlanations para encontrar um conjunto de características explicáveis e ao mesmo tempo, fornecer uma melhoria significativa no poder preditivo do modelo.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 426-431
Developmental Enhancement and Application of a Low-Cost Educational Robotics Kit: A Case Study Validation with a University Competition
This paper presents the latest developmental im-provements done on the URA low-cost robotic kit, which uses a open-source and low-cost robotics paradigm and was designed for general activities in educational robotics. Our kit is planned with the aim of facilitating the teaching of the basic concepts of programming, electronics, and mechanics. The current version (9.0) was redesigned with a focus on accessibility and intuitiveness in order to meet the demands of practical training in the educational context with limited infrastructure. The validation of the developments occurred in a mini-robotics competition held at the School of Science and Technology of the Federal University of Rio Grande do Norte, where freshmen undergraduate students assembled a robot with the kit and faced challenges inspired by the Brazilian Robotics Olympiad (OBR). The results pointed to the effectiveness of the kit in forming student teams, as well as its potential for replication in public schools and robotic events. This paper contributes with the development stages, the components, a simulation software, and programming methodology, pointing out the main difficulties encountered during implementation.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 426-431
Developmental Enhancement and Application of a Low-Cost Educational Robotics Kit: A Case Study Validation with a University Competition
This paper presents the latest developmental im-provements done on the URA low-cost robotic kit, which uses a open-source and low-cost robotics paradigm and was designed for general activities in educational robotics. Our kit is planned with the aim of facilitating the teaching of the basic concepts of programming, electronics, and mechanics. The current version (9.0) was redesigned with a focus on accessibility and intuitiveness in order to meet the demands of practical training in the educational context with limited infrastructure. The validation of the developments occurred in a mini-robotics competition held at the School of Science and Technology of the Federal University of Rio Grande do Norte, where freshmen undergraduate students assembled a robot with the kit and faced challenges inspired by the Brazilian Robotics Olympiad (OBR). The results pointed to the effectiveness of the kit in forming student teams, as well as its potential for replication in public schools and robotic events. This paper contributes with the development stages, the components, a simulation software, and programming methodology, pointing out the main difficulties encountered during implementation.
- Open Access
- 23 July 2025
- Published by Seven Events
- p. 111-128
HOW DATA MINING CAN BE USED TO IDENTIFY SPECIFIC CRIME PATTERNS IN THE CITY OF NATAL/RNBASED ON THE CHICAGO SCHOOL THEORY OF CRIMINOLOGY
Year after year, the increasing volume of data on crime in the state of Rio Grande do Norte (RN) exposes the fragility of the public security system and the new challenges. This prompts the use of advanced techniques to seek and extract valuable insights for the public security sector. This article proposes a modern perspective that combines data mining and machine learning algorithms with the CRISP-DM methodology. This arrangement aims to identify patterns and profiles of various crimes in Natal City, RN, enhancing the understanding of their occurrences by region and neighborhood within the city. The numbers indicate that this approach is not only effective as a tool for optimizing strategies but also has the ability to identify specific behavioral patterns. In the pursuit of crime prevention and reduction, this focus points to an effective remedy for addressing disputes related to criminality in RN.
- 23 July 2025
- Published by Seven Events
- p. 111-128
Open AccessHOW DATA MINING CAN BE USED TO IDENTIFY SPECIFIC CRIME PATTERNS IN THE CITY OF NATAL/RNBASED ON THE CHICAGO SCHOOL THEORY OF CRIMINOLOGY
Year after year, the increasing volume of data on crime in the state of Rio Grande do Norte (RN) exposes the fragility of the public security system and the new challenges. This prompts the use of advanced techniques to seek and extract valuable insights for the public security sector. This article proposes a modern perspective that combines data mining and machine learning algorithms with the CRISP-DM methodology. This arrangement aims to identify patterns and profiles of various crimes in Natal City, RN, enhancing the understanding of their occurrences by region and neighborhood within the city. The numbers indicate that this approach is not only effective as a tool for optimizing strategies but also has the ability to identify specific behavioral patterns. In the pursuit of crime prevention and reduction, this focus points to an effective remedy for addressing disputes related to criminality in RN.
- Open Access
- 3 July 2025
- Published by Brazilian Journals in REVISTA DELOS
- Vol. 18 (69) , e5808
Expanded analysis: requisitos para o desenvolvimento de um sistema multiplataforma de consulta de dados do censo da educação superior
Este trabalho apresenta a proposta de uma lista de requisitos para a construção de um sistema multiplataforma para trabalhar com dados do Censo da Educação Superior. Os dados na educação superior exigem ferramentas eficientes para análise e interpretação. Para tanto, a metodologia do estudo é qualitativa e será utilizado o Design Science Research, que permite o apropriado de modelos e consequente favorecimento da satisfação dos usuários. Como produto, o estudo oferecerá uma listagem de requisitos para instrumentos de consulta de dados da educação superior que favoreça processos de tomadas de decisão com informações estruturadas para a tomada de decisão visando o aprimoramento da transparência e o planejamento estratégico na educação superior por meio de uma interface acessível e lógica aos usuários.
- 3 July 2025
- Published by Brazilian Journals in REVISTA DELOS
- Vol. 18 (69) , e5808
Open AccessExpanded analysis: requisitos para o desenvolvimento de um sistema multiplataforma de consulta de dados do censo da educação superior
Este trabalho apresenta a proposta de uma lista de requisitos para a construção de um sistema multiplataforma para trabalhar com dados do Censo da Educação Superior. Os dados na educação superior exigem ferramentas eficientes para análise e interpretação. Para tanto, a metodologia do estudo é qualitativa e será utilizado o Design Science Research, que permite o apropriado de modelos e consequente favorecimento da satisfação dos usuários. Como produto, o estudo oferecerá uma listagem de requisitos para instrumentos de consulta de dados da educação superior que favoreça processos de tomadas de decisão com informações estruturadas para a tomada de decisão visando o aprimoramento da transparência e o planejamento estratégico na educação superior por meio de uma interface acessível e lógica aos usuários.
- Open Access
- 1 July 2025
- Published by Seven Events in ARACÊ
- Vol. 7 (7) , 38673-38691
COMO A MINERAÇÃO DE DADOS PODE SER UTILIZADA PARA IDENTIFICAR PADRÕES CRIMINAIS ESPECÍFICOS NA CIDADE DO NATAL/RN COM BASE NA TEORIA DA ESCOLA DE CHICAGO SOBRE A CRIMINOLOGIA
Ano após ano, o crescente volume de dados sobre a criminalidade no Estado do Rio Grande do Norte, expõe a fragilidade do sistema de segurança pública e os novos desafios a serem enfrentados, ensejando o uso de técnicas avançadas para buscar e retirar insights de grande valor para uso no setor de segurança pública. Este artigo sugere uma perspectiva moderna que une mineração de dados e algoritmos de machine learning e o CRISP-DM. Um arranjo que busca identificar padrões e perfis dos diversos crimes na Cidade do Natal/RN e aprimorar a compreensão de suas ocorrências, por região e bairro dentro da cidade. Os números indicam que essa abordagem não é eficaz apenas como ferramenta para aperfeiçoar estratégias, mas também tem a capacidade de identificar padrões comportamentais específicos, na busca da prevenção e diminuição dos crimes. Esse enfoque aponta para um remédio eficaz para tratar a criminalidade no RN.
- 1 July 2025
- Published by Seven Events in ARACÊ
- Vol. 7 (7) , 38673-38691
Open AccessCOMO A MINERAÇÃO DE DADOS PODE SER UTILIZADA PARA IDENTIFICAR PADRÕES CRIMINAIS ESPECÍFICOS NA CIDADE DO NATAL/RN COM BASE NA TEORIA DA ESCOLA DE CHICAGO SOBRE A CRIMINOLOGIA
Ano após ano, o crescente volume de dados sobre a criminalidade no Estado do Rio Grande do Norte, expõe a fragilidade do sistema de segurança pública e os novos desafios a serem enfrentados, ensejando o uso de técnicas avançadas para buscar e retirar insights de grande valor para uso no setor de segurança pública. Este artigo sugere uma perspectiva moderna que une mineração de dados e algoritmos de machine learning e o CRISP-DM. Um arranjo que busca identificar padrões e perfis dos diversos crimes na Cidade do Natal/RN e aprimorar a compreensão de suas ocorrências, por região e bairro dentro da cidade. Os números indicam que essa abordagem não é eficaz apenas como ferramenta para aperfeiçoar estratégias, mas também tem a capacidade de identificar padrões comportamentais específicos, na busca da prevenção e diminuição dos crimes. Esse enfoque aponta para um remédio eficaz para tratar a criminalidade no RN.
- 9 December 2024
- Published by Springer Nature
- p. 143-150
Designing a Resilient Hybrid-Modeled Digital Twin for Reliable Manufacturing Processes
- 9 December 2024
- Published by Springer Nature
- p. 143-150
Designing a Resilient Hybrid-Modeled Digital Twin for Reliable Manufacturing Processes
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 203-208
Musicalization of a Line-Following Robot for Educational Robotics Competitions
Educational Robotics competitions take place all over the world, with Robocup being the ultimate destination for most of these competitions. This is not different in Brazil. There are competitions for various educational levels, such as the Brazilian Robotics Competition (CBR) aimed at university students and the Brazilian Robotics Olympiad (OBR), intended for elementary and high school students. One of the categories in these competitions is the rescue category, where autonomous robots can be used to overcome challenges along a course, as well as rescue victims symbolized by objects. Various types of robots can be used in this category, utilizing wheels, chassis, claws, and sensors to assist in path detection, such as color, line, and proximity sensors. This category also has an extra challenge, where teams can present innovations in their own projects. In this sense, the project that performs best along the course and has the best innovation in the extra challenge earns points and can win the competition. One of the robots used in these competitions is from the project One Robot per Student (URA), of the Federal University of Rio Grande do Norte. The robot was designed both for use in competitions and for the training of teachers and teams. The use of music in Educational Robotics is a common approach and is known to be beneficial for education, but it is not as present in robotics competitions. Thus, as a matter of innovation, this work proposes to incorporate music into an autonomous robotics vehicle, in this case, the URA robot, making it a differentiating factor in educational robotics competitions. To achieve this, a buzzer actuator is attached to the URA vehicle's structure to produce sounds that signals the start and end of the vehicle's course in the rescue category challenge. These sounds are inserted into the programming through the frequencies of the melodies used in the URA vehicle's programming. The results of this work consider the effectiveness in competition environments within the rescue category in terms of functionality and performance.
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 203-208
Musicalization of a Line-Following Robot for Educational Robotics Competitions
Educational Robotics competitions take place all over the world, with Robocup being the ultimate destination for most of these competitions. This is not different in Brazil. There are competitions for various educational levels, such as the Brazilian Robotics Competition (CBR) aimed at university students and the Brazilian Robotics Olympiad (OBR), intended for elementary and high school students. One of the categories in these competitions is the rescue category, where autonomous robots can be used to overcome challenges along a course, as well as rescue victims symbolized by objects. Various types of robots can be used in this category, utilizing wheels, chassis, claws, and sensors to assist in path detection, such as color, line, and proximity sensors. This category also has an extra challenge, where teams can present innovations in their own projects. In this sense, the project that performs best along the course and has the best innovation in the extra challenge earns points and can win the competition. One of the robots used in these competitions is from the project One Robot per Student (URA), of the Federal University of Rio Grande do Norte. The robot was designed both for use in competitions and for the training of teachers and teams. The use of music in Educational Robotics is a common approach and is known to be beneficial for education, but it is not as present in robotics competitions. Thus, as a matter of innovation, this work proposes to incorporate music into an autonomous robotics vehicle, in this case, the URA robot, making it a differentiating factor in educational robotics competitions. To achieve this, a buzzer actuator is attached to the URA vehicle's structure to produce sounds that signals the start and end of the vehicle's course in the rescue category challenge. These sounds are inserted into the programming through the frequencies of the melodies used in the URA vehicle's programming. The results of this work consider the effectiveness in competition environments within the rescue category in terms of functionality and performance.
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 354-359
Educational Robotics Public Policies Implementation Issues
We aim to analyze the implementation process of a public policy provided by a municipal law enacted for the provision of educational robotics as a complementary activity in the education network of Natal, RN, Brazil. From the perspective of educational robotics trainees (undergraduate students), we highlight the construction potential and the challenges faced out by them using an action research paradigm done in six school units of the municipal network. The performance of the monitors has been systematized in the period from late 2021 to 2024, from which we aim to understand the law implementation. Activity reports were produced with the actions developed in each school in the mentioned period, to carry out a reflective data analysis in accordance with the legal provisions in the municipal regulatory framework. Therefore, the main contribution of the article is the analysis carried out on the experiences of implementing RE as a public policy, which can be used or replicated in other basic level education networks in Brazil. This includes a possible methodology that can serve as a support for other similar projects abroad. Above all, we noticed that planning must be rigorous, and strictly followed, otherwise it will demotivate teachers and school students. Also, the interference of external actors such as monitors and teachers from Higher Education Institutions is another factor that we analyze as having a strong influence in the continuity of such initiative.
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 354-359
Educational Robotics Public Policies Implementation Issues
We aim to analyze the implementation process of a public policy provided by a municipal law enacted for the provision of educational robotics as a complementary activity in the education network of Natal, RN, Brazil. From the perspective of educational robotics trainees (undergraduate students), we highlight the construction potential and the challenges faced out by them using an action research paradigm done in six school units of the municipal network. The performance of the monitors has been systematized in the period from late 2021 to 2024, from which we aim to understand the law implementation. Activity reports were produced with the actions developed in each school in the mentioned period, to carry out a reflective data analysis in accordance with the legal provisions in the municipal regulatory framework. Therefore, the main contribution of the article is the analysis carried out on the experiences of implementing RE as a public policy, which can be used or replicated in other basic level education networks in Brazil. This includes a possible methodology that can serve as a support for other similar projects abroad. Above all, we noticed that planning must be rigorous, and strictly followed, otherwise it will demotivate teachers and school students. Also, the interference of external actors such as monitors and teachers from Higher Education Institutions is another factor that we analyze as having a strong influence in the continuity of such initiative.
- Open Access
- 2 January 2024
- Published by Brazilian Journals in REVISTA DELOS
- Vol. 17 (51) , 98-117
Mineração de dados e análise no comportamento de vendas passagens fluviais na Amazônia
A cada dia, a crescente disponibilidade de dados coloca diante das empresas tanto oportunidades quanto desafios significativos, demandando a aplicação de técnicas avançadas para extrair insights valiosos. Este artigo propõe uma abordagem inovadora que une mineração de dados e algoritmos de machine learning, com destaque para o Random Forest e o Self-Organizing Map, utilizados de forma híbrida. Essa combinação visa não apenas identificar padrões e perfis de clientes em uma empresa de venda de passagens de barco, mas também aprimorar a satisfação do cliente ao compreender mais profundamente seus comportamentos. Os resultados destacam não apenas a eficácia dessa abordagem como uma ferramenta valiosa na otimização de estratégias, mas também sua capacidade de identificar padrões comportamentais específicos. Em um mercado altamente competitivo, onde a retenção dos clientes mais rentáveis se torna crucial, essa abordagem se revela como uma resposta eficiente para enfrentar os desafios de um ambiente comercial dinâmico e altamente competitivo.
- 2 January 2024
- Published by Brazilian Journals in REVISTA DELOS
- Vol. 17 (51) , 98-117
Open AccessMineração de dados e análise no comportamento de vendas passagens fluviais na Amazônia
A cada dia, a crescente disponibilidade de dados coloca diante das empresas tanto oportunidades quanto desafios significativos, demandando a aplicação de técnicas avançadas para extrair insights valiosos. Este artigo propõe uma abordagem inovadora que une mineração de dados e algoritmos de machine learning, com destaque para o Random Forest e o Self-Organizing Map, utilizados de forma híbrida. Essa combinação visa não apenas identificar padrões e perfis de clientes em uma empresa de venda de passagens de barco, mas também aprimorar a satisfação do cliente ao compreender mais profundamente seus comportamentos. Os resultados destacam não apenas a eficácia dessa abordagem como uma ferramenta valiosa na otimização de estratégias, mas também sua capacidade de identificar padrões comportamentais específicos. Em um mercado altamente competitivo, onde a retenção dos clientes mais rentáveis se torna crucial, essa abordagem se revela como uma resposta eficiente para enfrentar os desafios de um ambiente comercial dinâmico e altamente competitivo.