Skip to main content
Understanding the core function of the brain is one the major challenges of our times. In the areas of neuroscience and education, several new studies try to correlate the learning difficulties faced by children and youth with behavioral... more
Understanding the core function of the brain is one the major challenges of our times. In the areas of neuroscience and education, several new studies try to correlate the learning difficulties faced by children and youth with behavioral and social problems. This work aims to present the challenges and opportunities of computational neuroscience research, with the aim of detecting people with learning disorders. We present a line of investigation based on the key areas: neuroscience, cognitive sciences and computer science, which considers young people between nine and eighteen years of age, with or without a learning disorder. The adoption of neural networks reveals consistency in dealing with pattern recognition problems and they are shown to be effective for early detection in patients with these disorders. We argue that computational neuroscience can be used for identifying and analyzing young Brazilian people with several cognitive disorders.
Scientific Workflows are abstractions used to model in silico scientific experiments. Cloud environments are still incipient in collecting and recording prospective and retrospective provenance. This paper presents an approach to support... more
Scientific Workflows are abstractions used to model in silico scientific experiments. Cloud environments are still incipient in collecting and recording prospective and retrospective provenance. This paper presents an approach to support collecting metadata provenance of in silico scientific experiments executed in public clouds. The strategy was implemented as a distributed and modular architecture named Matriohska. This paper also presents a provenance data model compatible with PROV specification. We also show preliminary results that describe how provenance metadata was captured from the components running in the cloud.
Research Interests:
Download (.pdf)
Download (.pdf)
ABSTRACT Scientific workflows have been established as a ordinary approach adopted by researchers to create simplified views of complex scientific processes. Science portals are a way to simplify the management of large in silico... more
ABSTRACT Scientific workflows have been established as a ordinary approach adopted by researchers to create simplified views of complex scientific processes. Science portals are a way to simplify the management of large in silico scientific experiments, they aggregate data from different sources and by providing a set of pre-designed analyses. However, such portals are often built manually, and are not flexible enough to support sharing, reusing and executing workflows enacted by SWfMS. In this paper we describe VTPortal, a science portal that combines a set of tools and an infrastructure for providing a collaborative environment for scientists. VTPortal takes into account the requirements of computational scientists, such as accessing high-performance computers and manipulating large amounts of data.
Page 1. Previsão de Estudantes com Risco de Evasão Utilizando Técnicas de Mineração de Dados Laci Mary Barbosa Manhães 1 , Sérgio Manuel Serra da Cruz 12 , Raimundo J. Macário Costa 1 , Jorge Zavaleta 1 , Geraldo Zimbrão 1 ...
Download (.pdf)
This paper uses data mining techniques to indentify key variables related with students failures in completing their undergraduate studies. In our approach, classification analysis is used to manipulate academic data of students of the... more
This paper uses data mining techniques to indentify key variables related with students failures in completing their undergraduate studies. In our approach, classification analysis is used to manipulate academic data of students of the largest Brazilian Federal University. Differently from other works, our research shows that even analyzing three different classes of students it was possible to have a global precision above 80%. The Naïve Bayes model was used to visualize the key variables used to separate the distinct classes of students. Resumo. Este artigo utiliza técnicas de mineração de dados para identificar problemas relacionados com alunos que não conseguem completar os seus cursos de graduação. Nessa abordagem, a classificação manipula informações acadêmicas de alunos oriundos de uma grande Universidade Federal Brasileira. Muitas técnicas de mineração de dados foram avaliadas em função da acurácia obtida quando aplicadas a um conjunto de dados dos estudantes universitários....
Download (.pdf)
This paper aims to present a modular tool based on computational intelligence techniques that can support specialists in the detection, evaluation, intervention in the diagnosis of dyslexia. The evaluation module is based on unsupervised... more
This paper aims to present a modular tool based on computational intelligence techniques that can support specialists in the detection, evaluation, intervention in the diagnosis of dyslexia. The evaluation module is based on unsupervised neural networks to detect and to classify dyslexia patterns and validated using a sample of 52 students of 09-18 years old. The intervention module uses fuzzy logic to deduce and to classify the different degrees of dyslexia and suggest the most appropriate learning strategy to be applied to each individual and is under development and validation.
Download (.pdf)
Este trabalho apresenta os resultados obtidos na implementação computacional de um modelo de agregação e consolidação de dados distribuídos em apoio à gestão de saúde. Pretende-se que este modelo de agregação sirva de suporte e fonte de... more
Este trabalho apresenta os resultados obtidos na implementação computacional de um modelo de agregação e consolidação de dados distribuídos em apoio à gestão de saúde. Pretende-se que este modelo de agregação sirva de suporte e fonte de dados aos pesquisadores da área de saúde com vistas ao planejamento de ações de saúde, em especial em instâncias municipais. Com este propósito, é implementado um processo de obtenção e elaboração automática de dados. Apresentamos o processo de definição deste modelo de obtenção das informações, incluindo o tipo de suporte, o processo da obtenção dos registros das produções, o modelo de vinculação às redes sociais e a produção acadêmica. A partir do modelo aqui apresentado foram gerados 91 e-books que estão disponíveis, em acesso aberto. ABSTRACT This paper presents the computational implementation of a data distributed fusion and consolidation model in support of health management. It is intended that this aggregation model work as a support and sou...
RESUMO A compreensão do funcionamento do cérebro se constitui em um dos grandes desafios do momento. Na área de educação, novos estudos buscam relacionar as dificuldades de aprendizagem enfrentadas por crianças e jovens com problemas... more
RESUMO A compreensão do funcionamento do cérebro se constitui em um dos grandes desafios do momento. Na área de educação, novos estudos buscam relacionar as dificuldades de aprendizagem enfrentadas por crianças e jovens com problemas comportamentais e sociais. Este trabalho tem por objetivo apresentar um ambiente computacional inteligente baseado em neurociência computacional para detectar pessoas com possíveis riscos de dislexia ou distúrbios de aprendizagem. A investigação foi avaliada através de um estudo com pessoas na faixa etária de 9 a 18 anos com ou sem o diagnóstico de dislexia. A adoção da técnica de redes neuronais demonstrou consistência ao lidar com os problemas de reconhecimento de padrão e indicam que para fins de rastreio se mostram eficientes no diagnóstico precoce em escolares. A compreensão do funcionamento do cérebro se constitui em um dos grandes desafios do momento. A neurociência contribui no objetivo de estudar e analisar o sistema nervoso central (SNC) dos s...
Download (.pdf)
Download (.pdf)
Download (.pdf)
Download (.pdf)
Download (.pdf)
Understanding the core function of the brain is one the major challenges of our times. In the areas of neuroscience and education, several new studies try to correlate the learning difficulties faced by children and youth with behavioral... more
Understanding the core function of the brain is one the major challenges of our times. In the areas of neuroscience and education, several new studies try to correlate the learning difficulties faced by children and youth with behavioral and social problems. This work aims to present the challenges and opportunities of computational neuroscience research, with the aim of detecting people with learning disorders. We present a line of investigation based on the key areas: neuroscience, cognitive sciences and computer science, which considers young people between nine and eighteen years of age, with or without a learning disorder. The adoption of neural networks reveals consistency in dealing with pattern recognition problems and they are shown to be effective for early detection in patients with these disorders. We argue that computational neuroscience can be used for identifying and analyzing young Brazilian people with several cognitive disorders.
Research Interests:
Download (.pdf)
Page 1. Revista Tecer - Belo Horizonte – vol. 4, nº 6, maio 2011 41 Abordagem tecnológica para rastreio de pessoas com dislexia Raimundo José Macário Costa Sérgio Manuel Serra da Cruz Jorge Zavaleta Luís Alfredo Vidal Filho Renata... more
Page 1. Revista Tecer - Belo Horizonte – vol. 4, nº 6, maio 2011 41 Abordagem tecnológica para rastreio de pessoas com dislexia Raimundo José Macário Costa Sérgio Manuel Serra da Cruz Jorge Zavaleta Luís Alfredo Vidal Filho Renata Mousinho ...
ABSTRACT Scientific Workflows are abstractions used to model in silico scientific experiments. Cloud environments are still incipient in collecting and recording prospective and retrospective provenance. This paper presents an approach to... more
ABSTRACT Scientific Workflows are abstractions used to model in silico scientific experiments. Cloud environments are still incipient in collecting and recording prospective and retrospective provenance. This paper presents an approach to support collecting metadata provenance of in silico scientific experiments executed in public clouds. The strategy was implemented as a distributed and modular architecture named Matriohska. This paper also presents a provenance data model compatible with PROV specification. We also show preliminary results that describe how provenance metadata was captured from the components running in the cloud.
... 93-126. [4] JV Dempsey, LL Haynes, BA Lucassen, and MS Casey, “Forty simple computer games and what they could mean to educators”, Simulations & Gaming, Vol. 33 No. ... 71. [8] Manual dePsicologia Aplicada, WISC – escala de... more
... 93-126. [4] JV Dempsey, LL Haynes, BA Lucassen, and MS Casey, “Forty simple computer games and what they could mean to educators”, Simulations & Gaming, Vol. 33 No. ... 71. [8] Manual dePsicologia Aplicada, WISC – escala de inteligência Wechsler para crianças. ...
Jogos Inteligentes são Educacionais? Leila ANDRADE1 e 2, Jorge ZAVALETA2, Francine VAZ2, Cabral LIMA2, Catia ARAUJO2, Adriana SOARES2 ... São Paulo: Pioneira, 1979. [Russel e Norvig, 1995] RUSSELL, Stuart; NORVIG, Peter. ... Editora... more
Jogos Inteligentes são Educacionais? Leila ANDRADE1 e 2, Jorge ZAVALETA2, Francine VAZ2, Cabral LIMA2, Catia ARAUJO2, Adriana SOARES2 ... São Paulo: Pioneira, 1979. [Russel e Norvig, 1995] RUSSELL, Stuart; NORVIG, Peter. ... Editora Martins Fontes. São Paulo, 2000. ...
Download (.pdf)
ABSTRACT Scientific workflows have been established as a ordinary approach adopted by researchers to create simplified views of complex scientific processes. Science portals are a way to simplify the management of large in silico... more
ABSTRACT Scientific workflows have been established as a ordinary approach adopted by researchers to create simplified views of complex scientific processes. Science portals are a way to simplify the management of large in silico scientific experiments, they aggregate data from different sources and by providing a set of pre-designed analyses. However, such portals are often built manually, and are not flexible enough to support sharing, reusing and executing workflows enacted by SWfMS. In this paper we describe VTPortal, a science portal that combines a set of tools and an infrastructure for providing a collaborative environment for scientists. VTPortal takes into account the requirements of computational scientists, such as accessing high-performance computers and manipulating large amounts of data.
Download (.pdf)