51 resultados para learning with errors
em Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho"
Resumo:
Concept drift is a problem of increasing importance in machine learning and data mining. Data sets under analysis are no longer only static databases, but also data streams in which concepts and data distributions may not be stable over time. However, most learning algorithms produced so far are based on the assumption that data comes from a fixed distribution, so they are not suitable to handle concept drifts. Moreover, some concept drifts applications requires fast response, which means an algorithm must always be (re) trained with the latest available data. But the process of labeling data is usually expensive and/or time consuming when compared to unlabeled data acquisition, thus only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are also based on the assumption that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenge in machine learning. Recently, a particle competition and cooperation approach was used to realize graph-based semi-supervised learning from static data. In this paper, we extend that approach to handle data streams and concept drift. The result is a passive algorithm using a single classifier, which naturally adapts to concept changes, without any explicit drift detection mechanism. Its built-in mechanisms provide a natural way of learning from new data, gradually forgetting older knowledge as older labeled data items became less influent on the classification of newer data items. Some computer simulation are presented, showing the effectiveness of the proposed method.
Resumo:
Concept drift, which refers to non stationary learning problems over time, has increasing importance in machine learning and data mining. Many concept drift applications require fast response, which means an algorithm must always be (re)trained with the latest available data. But the process of data labeling is usually expensive and/or time consuming when compared to acquisition of unlabeled data, thus usually only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are based on assumptions that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenging task in machine learning. Recently, a particle competition and cooperation approach has been developed to realize graph-based semi-supervised learning from static data. We have extend that approach to handle data streams and concept drift. The result is a passive algorithm which uses a single classifier approach, naturally adapted to concept changes without any explicit drift detection mechanism. It has built-in mechanisms that provide a natural way of learning from new data, gradually "forgetting" older knowledge as older data items are no longer useful for the classification of newer data items. The proposed algorithm is applied to the KDD Cup 1999 Data of network intrusion, showing its effectiveness.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
Pós-graduação em Odontologia Preventiva e Social - FOA
Resumo:
Studies show the positive effects that video games can have on student performance and attitude towards learning. In the past few years, strategies have been generated to optimize the use of technological resources with the aim of facilitating widespread adoption of technology in the classroom. Given its low acquisition and maintenance costs, the interpersonal computer allows individual interaction and simultaneous learning with large groups of students. The purpose of this work was to compare arithmetical knowledge acquired by third-grade students through the use of game-based activities and non-game-based activities using an interpersonal computer, with knowledge acquired through the use of traditional paper-and-pencil activities, and to analyze their impact in various socio-cultural contexts. To do this, a quasi-experimental study was conducted with 271 students in three different countries (Brazil, Chile, and Costa Rica), in both rural and urban schools. A set of educational games for practising arithmetic was developed and tested in six schools within these three countries. Results show that there were no significant differences (ANCOVA) in the learning acquired from game-based vs. non-game-based activities. However, both showed a significant difference when compared with the traditional method. Additionally, both groups using the interpersonal computer showed higher levels of student interest than the traditional method group, and these technological methods were seen to be especially effective in increasing learning among weaker students.
Resumo:
This article presents some of the results of a qualitative research project about the influences of the pedagogic strategies used by a mediator (graduate student in applied linguistics) in the supervision process of a Teletandem partner (undergraduate student in languages) on her pedagogical practice. It was done within the project Teletandem Brazil: foreign language for all. Based on the reflective teaching paradigm and collaborative language learning, with special emphasis on tandem learning, we analyzed the contributions of the collaborative relationship established between the graduate student and the student-teacher in her first teaching experience. The results bring about implications for the field of language teacher education in a perspective of education within practice, evidencing the experience of collaborative learning in teletandem as an opportunity for reflective teacher education of pre-service teachers.
Resumo:
In the context of medical school instruction, the segmented approach of a focus on specialties and excessive use of technology seem to hamper the development of the professional-patient relationship and an understanding of the ethics of this relationship. The real world presents complexities that require multiple approaches. Engagement in the community where health competence is developed allows extending the usefulness of what is learned. Health services are spaces where the relationship between theory and practice in health care are real and where the social role of the university can be revealed. Yet some competencies are still lacking and may require an explicit agenda to enact. Ten topics are presented for focus here: environmental awareness, involvement of students in medical school, social networks, interprofessional learning, new technologies for the management of care, virtual reality, working with errors, training in management for results, concept of leadership, and internationalization of schools. Potential barriers to this agenda are an underinvestment in ambulatory care infrastructure and community-based health care facilities, as well as in information technology offered at these facilities; an inflexible departmental culture; and an environment centered on a discipline-based medical curriculum.
Resumo:
A estimativa do diâmetro de lesões de cancro cítrico é uma das principais técnicas usadas na avaliação da interação entre isolados x genótipos, na avaliação da resistência varietal e no estudo de aspectos epidemiológicos. No entanto, inexistem informações a respeito do tamanho da amostra para uma adequada quantificação da doença, considerando o diâmetro de lesões. Nesse sentido, o presente trabalho teve como objetivo a determinação do número de amostras para mensurar o diâmetro médio de lesões de cancro cítrico. Foram considerados como fontes de variação três genótipos do hospedeiro, dois métodos de inoculação de Xanthomonas citri subsp. citri e dois avaliadores. As avaliações dos diâmetros foram realizadas considerando ou não o halo amarelo ao redor do tecido necrosado, quando presente. Estimativas do diâmetro de lesões com erros na média inferiores a 3% são impraticáveis em razão do tamanho excessivo da amostra (mais que quarenta lesões/planta). Menores erros na média ocorreram nas estimativas considerando somente o tecido necrosado. Estimativas precisas do diâmetro de lesões, com erros na média inferiores a 10%, podem ser obtidas com tamanhos de amostras entre cinco e quinze lesões/planta, independentemente do genótipo e da idade das lesões.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
Este artigo apresenta os resultados de um estudo qualitativo realizado no projeto Teletandem Brasil sobre os processos de supervisão e estratégias pedagógicas utilizadas por uma mediadora no processo de mediação de um par interagente. Baseando-nos no paradigma da formação reflexiva, na teoria sociocultural de Vygotsky e na aprendizagem colaborativa de línguas, com especial ênfase para o regime tandem, analisamos as contribuições da relação colaborativa formada entre uma aluna de pós-graduação e uma aluna-professora em formação. Os resultados trazem implicações para o campo da formação de professores de línguas, em relação a uma perspectiva de formação na prática, evidenciando a experiência de ensino e aprendizagem colaborativos no Teletandem como uma oportunidade para a formação reflexiva do aluno-professor, o interagente, e também do futuro formador de professores, o mediador.
Multivariate quality control studies applied to Ca(II) and Mg(II) determination by a portable method
Resumo:
A portable or field test method for simultaneous spectrophotometric determination of calcium and magnesium in water using multivariate partial least squares (PLS) calibration methods is proposed. The method is based on the reaction between the analytes and methylthymol blue at pH 11. The spectral information was used as the X-block, and the Ca(II) and Mg(II) concentrations obtained by a reference technique (ICP-AES) were used as the Y-block. Two series of analyses were performed, with a month's difference between them. The first series was used as the calibration set and the second one as the validation set. Multivariate statistical process control (MSPC) techniques, based on statistics from principal component models, were used to study the features and evolution with time of the spectral signals. Signal standardization was used to correct the deviations between series. Method validation was performed by comparing the predictions of the PLS model with the reference Ca(II) and Mg(II) concentrations determined by ICP-AES using the joint interval test for the slope and intercept of the regression line with errors in both axes. (C) 1998 John Wiley & Sons, Ltd.
Resumo:
Pós-graduação em Televisão Digital: Informação e Conhecimento - FAAC
Resumo:
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Resumo:
Pós-graduação em Educação - FFC
Resumo:
Pós-graduação em Odontologia - FOA