Uso de confiabilidade na rotulação de exemplos em problemas de classificação multirrótulo com aprendizado semissupervisionado

The techniques of Machine Learning are applied in classification tasks to acquire knowledge through a set of data or information. Some learning methods proposed in literature are methods based on semissupervised learning; this is represented by small percentage of labeled data (supervised learning) combined with a quantity of label and non-labeled examples (unsupervised learning) during the training phase, which reduces, therefore, the need for a large quantity of labeled instances when only small dataset of labeled instances is available for training. A commom problem in semi-supervised learning is as random selection of instances, since most of paper use a random selection technique which can cause a negative impact. Much of machine learning methods treat single-label problems, in other words, problems where a given set of data are associated with a single class; however, through the requirement existent to classify data in a lot of domain, or more than one class, this classification as called multi-label classification. This work presents an experimental analysis of the results obtained using semissupervised learning in troubles of multi-label classification using reliability parameter as an aid in the classification data. Thus, the use of techniques of semissupervised learning and besides methods of multi-label classification, were essential to show the results

Investigando a combinação de técnicas de aprendizado semissupervisionado e classificação hierárquica multirrótulo

Data classification is a task with high applicability in a lot of areas. Most methods for treating classification problems found in the literature dealing with single-label or traditional problems. In recent years has been identified a series of classification tasks in which the samples can be labeled at more than one class simultaneously (multi-label classification). Additionally, these classes can be hierarchically organized (hierarchical classification and hierarchical multi-label classification). On the other hand, we have also studied a new category of learning, called semi-supervised learning, combining labeled data (supervised learning) and non-labeled data (unsupervised learning) during the training phase, thus reducing the need for a large amount of labeled data when only a small set of labeled samples is available. Thus, since both the techniques of multi-label and hierarchical multi-label classification as semi-supervised learning has shown favorable results with its use, this work is proposed and used to apply semi-supervised learning in hierarchical multi-label classication tasks, so eciently take advantage of the main advantages of the two areas. An experimental analysis of the proposed methods found that the use of semi-supervised learning in hierarchical multi-label methods presented satisfactory results, since the two approaches were statistically similar results

Previsão do índice bovespa por meio de redes neurais artificiais: uma análise comparada aos métodos tradicionais de séries de tempo

Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model

Modelos de dispositivos de microondas e ópticos através de redes neurais artificiais de alimentação direta

This dissertation contributes for the development of methodologies through feed forward artificial neural networks for microwave and optical devices modeling. A bibliographical revision on the applications of neuro-computational techniques in the areas of microwave/optical engineering was carried through. Characteristics of networks MLP, RBF and SFNN, as well as the strategies of supervised learning had been presented. Adjustment expressions of the networks free parameters above cited had been deduced from the gradient method. Conventional method EM-ANN was applied in the modeling of microwave passive devices and optical amplifiers. For this, they had been proposals modular configurations based in networks SFNN and RBF/MLP objectifying a bigger capacity of models generalization. As for the training of the used networks, the Rprop algorithm was applied. All the algorithms used in the attainment of the models of this dissertation had been implemented in Matlab

Previsão do índice bovespa por meio de redes neurais artificiais: uma análise comparada aos métodos tradicionais de séries de tempo

Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model

Estágio Supervisionado na formação de professores em serviço dos anos iniciais do ensino fundamental

This report aims to present the results of research on the possibilities and limits of Supervised Traineeship in a the course of Training of Teachers in Service of the early years of primary education conducted through an agreement between the Federal University of Piauí and the Teresina city council (PI) UFPI/PMT, offers subsidies to teaching practices that address the reality of students who live, work and/or studying in school located in rural areas. The research is based on a theoretical and methodological approach that places the object of study in the critical perspective and using procedures such as: literature (bibliographical research), documentary analysis, semi-structured interview and photographic record. We researched 10 (ten) teachers of the course promoted by UFPI graduating in 2006 and who develop teaching activities in schools located in rural areas in the city of Teresina-PI, 02 (two) teachers who are trainee supervisor of UFPI and the coordinator of the course. The analysis evidenced that the object, historically, educational policies have changed, but did not break the traditional patterns of teacher education, enough, especially in the curricular proposals in light of the changes required by the information and global society. We also note that in the supervised training, the process of critical reflection on teaching practice was made possible, in part, and raised the reconstruction of specific knowledge to make teaching in order to conduct the teaching-learning process articulating different realities of primary teaching and facing situations due to conservative practices in relation to education in rural areas. It contributed, therefore, this curricular component to resize, in part, the pedagogical action of the teachers graduated. The study also drives the research toward to similar studies in the perspective of overcoming the current model of teacher training that does not correspond to the demands of society in change and to the crisis of capital, with immeasurable consequences in the workplace. The contemporary claims for a University to form professionals capable of directing the destiny of society, where teacher training is a social and political demands on which is co-responsible

O estágio supervisionado como contexto de formação docente específica para educação infantil: o que dizem os formandos sobre suas aprendizagens?

Kindergarten teachers training gains the spotlight with the passing of Law number 9.394/96 (Guidelines and Basis Law) that defines this segment as the initial step of basic education, with pedagogical function. In this spectrum, the discussion about teacher training unravels to ensure social quality to education as well as the teacher s specificities towards child singularities. Adding to that, the growing propagation of Pedagogy in an undergraduate level, given that such course has been continually transformed by the National Curriculum Guidelines for Pedagogy (2006), highlighting the addition of curriculum components that are specific to upbringing. The complex debate circa kindergarten teachers training has advancements and hardships that need to be unveiled in order to improve both formation and social quality of education in the 0- 5 years old range. This investigation inserts itself in said context and aims to analyze which knowledge, specific to kindergarten teaching are constructed, according to undergraduate trainees, in Pedagogy s supervised internship. The study was conducted alongside the discipline: Supervised Internship in Child Education ministered by the Advanced Campus of Rio Grande do Norte s State University s in the municipality of Patu-RN (CAP-UERN-Patu) and was conducted through 2012 by accompanying four undergraduate interns. We first assumed that the development of teaching knowledge is a complex process of appropriation of cultural-social practices and is symbolically mediated by interactions that occur in the formation context, and the supervised internship can be understood as a space for the articulation and enlargement of theoretical and practical knowledge, directly related to the specificities of child education. The theoretical-methodological foundation was based upon the historiccultural approach of L. S. Vygotsky and M. Bakhtin s dialogism on human sciences research, as well as his postulates on learning and developmental processes, conceived as both essentially social and discursive. The investigation approached the principles of the qualitative perspective and to the construction and analysis of data, involved documental analysis and, specially, semi-structured interviews, both individual and collective, whose fundamental premise was the production-comprehension of meanings in a dialogical perspective. The participants texts/speeches produced a synthesis that points to the occurrence, within the supervised internship at CAP/UERN, of internalization/appropriation processes and, as such, of formulation of meanings that are pertinent to child education: child, childhood, kindergarten and teacher signification and this stage s specific teaching knowledge. It stood out that the internship, alongside other curriculum components, is, in fact, one of the primeval formation environment for the teachers, in which the interns interact with their colleagues, supervisor professor, collaborator professor, and of course, the children to construct their erudition. Such interactions allow the undergraduate interns to develop attitudes and procedures to reflect on what they know, what they ve done and what they can achieve. We have concluded that the undergraduate internship can constitute itself as an articulatorconsolidator environment in the future teacher s formation process and, since well oriented, can provide the effective initiation, not only to the practice, but to the praxis as a movement of non dissociability between theory and practice

Preceptoria em profissionais de saúde: viabilizando meios para uma formação pedagógica em hospitais públicos em Natal-RN

This study aimed to construct and carry out a distance course of pedagogical training for health professional performing preceptorship functions in public health institutions. The preceptorship in health is a pedagogical practice that occurs in workplace, led by assistance professionals with teaching position or not, where the vast majority of these acts intuitively, reproducing their own training, confusing transmitting information with education. These preceptors often do not dominate the pedagogical knowledge, necessary for the organization of training activities, such as the various teaching-learning processes and the different assessment types. Student supervision is essential in the training process of students in the health field, and on the occasion of supervised internships that the teaching-learning process is based on practical experience with participation in real life situations and professional work. It was realized an exploratory study, descriptive with qualitative approach, with the development of tutoring teaching course in health as final product. Applied semi structured research instrument from may to july 2014. It were evaluated 162 health professionals who perform the preceptorship, which made it possible define the preceptor's profile and identify the educational requirements related to the educational process, which justified the construction of the program content and the professionals’ perception analysis about preceptorship through identification of three categories: clinic knowledge valuation; valuation of professional orientation and valuation of professional future. The course was available on distance mode through Moodle platform with forty hours of work load from October to November 2014. With the aim of capacitate the health professionals to development of necessary abilities and skills to tutoring performance through thoughts about tutoring concepts, professional training within the curricular guidelines and SUS precepts, the role of health professionals as educators, application of active teaching methodologies, and evaluation methods. The applications were done online through the provided link; 300 vacancies offered, 243 professionals applied, chosen 134 that works on tutoring, where 49 represented professionals that works on the location of the study. The course lasted 45 days, and counted with tutors responsible to interact and evaluate the students. 28 professionals joined the course, 12 concluded. Opportunities were identified to stimulate the involvement, however the professionals’ satisfaction shows that, make an investment on tutors education, starting from the Permanent Education precepts, will provide a bigger appropriation of the knowledge to the education and therefore the awareness of their role as an educator on work ambit, proportioning essential tools to tutors act while enabler of integration between theory and practice and result better teaching-learning process.