Otimização em comitês de classificadores: uma abordagem baseada em filtro para seleção de subconjuntos de atributos

Traditional applications of feature selection in areas such as data mining, machine learning and pattern recognition aim to improve the accuracy and to reduce the computational cost of the model. It is done through the removal of redundant, irrelevant or noisy data, finding a representative subset of data that reduces its dimensionality without loss of performance. With the development of research in ensemble of classifiers and the verification that this type of model has better performance than the individual models, if the base classifiers are diverse, comes a new field of application to the research of feature selection. In this new field, it is desired to find diverse subsets of features for the construction of base classifiers for the ensemble systems. This work proposes an approach that maximizes the diversity of the ensembles by selecting subsets of features using a model independent of the learning algorithm and with low computational cost. This is done using bio-inspired metaheuristics with evaluation filter-based criteria

MoSAC: um módulo para seleção de configurações arquiteturais baseado em critérios de QoS no contexto de sistema distribuídos autoadaptativos

The process for choosing the best components to build systems has become increasingly complex. It becomes more critical if it was need to consider many combinations of components in the context of an architectural configuration. These circumstances occur, mainly, when we have to deal with systems involving critical requirements, such as the timing constraints in distributed multimedia systems, the network bandwidth in mobile applications or even the reliability in real-time systems. This work proposes a process of dynamic selection of architectural configurations based on non-functional requirements criteria of the system, which can be used during a dynamic adaptation. This proposal uses the MAUT theory (Multi-Attribute Utility Theory) for decision making from a finite set of possibilities, which involve multiple criteria to be analyzed. Additionally, it was proposed a metamodel which can be used to describe the application s requirements in terms of the non-functional requirements criteria and their expected values, to express them in order to make the selection of the desired configuration. As a proof of concept, it was implemented a module that performs the dynamic choice of configurations, the MoSAC. This module was implemented using a component-based development approach (CBD), performing a selection of architectural configurations based on the proposed selection process involving multiple criteria. This work also presents a case study where an application was developed in the context of Digital TV to evaluate the time spent on the module to return a valid configuration to be used in a middleware with autoadaptative features, the middleware AdaptTV

Seleção de atributos em comitês de classificadores utilizando algoritmos genéticos

Classifier ensembles are systems composed of a set of individual classifiers and a combination module, which is responsible for providing the final output of the system. In the design of these systems, diversity is considered as one of the main aspects to be taken into account since there is no gain in combining identical classification methods. The ideal situation is a set of individual classifiers with uncorrelated errors. In other words, the individual classifiers should be diverse among themselves. One way of increasing diversity is to provide different datasets (patterns and/or attributes) for the individual classifiers. The diversity is increased because the individual classifiers will perform the same task (classification of the same input patterns) but they will be built using different subsets of patterns and/or attributes. The majority of the papers using feature selection for ensembles address the homogenous structures of ensemble, i.e., ensembles composed only of the same type of classifiers. In this investigation, two approaches of genetic algorithms (single and multi-objective) will be used to guide the distribution of the features among the classifiers in the context of homogenous and heterogeneous ensembles. The experiments will be divided into two phases that use a filter approach of feature selection guided by genetic algorithm

Uma ferramenta de análise automatizada de técnicas de seleção de testes de regressão baseada em mineração de repositórios de software

The main goal of Regression Test (RT) is to reuse the test suite of the latest version of a software in its current version, in order to maximize the value of the tests already developed and ensure that old features continue working after the new changes. Even with reuse, it is common that not all tests need to be executed again. Because of that, it is encouraged to use Regression Tests Selection (RTS) techniques, which aims to select from all tests, only those that reveal faults, this reduces costs and makes this an interesting practice for the testing teams. Several recent research works evaluate the quality of the selections performed by RTS techniques, identifying which one presents the best results, measured by metrics such as inclusion and precision. The RTS techniques should seek in the System Under Test (SUT) for tests that reveal faults. However, because this is a problem without a viable solution, they alternatively seek for tests that reveal changes, where faults may occur. Nevertheless, these changes may modify the execution flow of the algorithm itself, leading some tests no longer exercise the same stretch. In this context, this dissertation investigates whether changes performed in a SUT would affect the quality of the selection of tests performed by an RTS, if so, which features the changes present which cause errors, leading the RTS to include or exclude tests wrongly. For this purpose, a tool was developed using the Java language to automate the measurement of inclusion and precision averages achieved by a regression test selection technique for a particular feature of change. In order to validate this tool, an empirical study was conducted to evaluate the RTS technique Pythia, based on textual differencing, on a large web information system, analyzing the feature of types of tasks performed to evolve the SUT

Fatores de riscos cardiovasculares em adolescentes da cidade do Natal-RN

Estudos têm demonstrado elevada prevalência de fatores de risco cardiovascular em adolescentes a redor do mundo, entretanto, é possível que esses fatores de risco se comportem de modo diferente em diferentes localidades. Objetivos: Examinar a prevalência do sobrepeso/obesidade e sua associação com outros fatores de risco cardiovasculares, em adolescentes, da cidade do Natal-Brasil. Métodos: Estudo observacional de delineamento transversal, realizado em 626 adolescentes (10 a 19 anos) de ambos os sexos. Foram estudadas as variáveis: peso, idade, gênero, cor, escolaridade, renda familiar, hábitos de vida, história familiar, peso, estatura, índice de massa corporal, relação cintura/quadril, pressão arterial, perfil lipídico, Glicose e Insulina de jejum e pós Dextrosol. Resultados: Foram avaliados 273 (43,6%) adolescentes do sexo masculino e 353 (56,4%) feminino. 26,4% dos adolescentes apresentaram sobrepeso/obesidade. A presença de obesidade familiar foi relatada por mais de 30 % da nossa amostra. Na análise de regressão logística múltipla; Idade, renda familiar, percentil de pressão sistólica, história familiar de hipertensão e obesidade, triglicerídeos, HDL colesterol, e HOMA RI mostraram-se associados com sobrepeso/obesidade. A relação cintura quadril apresentou-se mais elevada nas mulheres, e encontramos 10,9 % dos adolescentes com percentil de pressão sistólica (PAS) 95, e 7,4 % com percentil de pressão diastólica (PAD) 95. As dosagens de triglicerídeos, colesterol HDL e HOMA-RI alterados foram mais prevalentes nos que apresentavam IMC aumentado. As alterações do xi colesterol total, triglicerídios, glicemia pós dextrosol e HOMA teste, tiveram maior prevalência no gênero feminino. Na regressão logística binária, foram observadas associações do sobrepeso / obesidade com idade; OR 0,85, IC de 95% (0,78-0,92); p<0,001, pressão arterial sistólica; OR 2,65, IC de 95% (1,18- 5,94); p< 0,020, renda familiar; OR 2,34, IC de 95% (1,53-3,58); p< 0,001, história familiar de hipertensão arterial; OR 1,76, IC de 95% (1,15-2,71); p< 0,009, história familiar de obesidade; OR 1,50, IC de 95% (1,09-2,27); p< 0,04, aumento dos trigliceridios; OR 2,74, IC de 95% (1,69-4,43); p< 0,001, redução do colesterol; HDL OR 0,58, IC de 95% (0,38-0,87); p< 0,009 e o aumento do HOMA OR 3,16, IC de 95% (1,64 - 6,02); p<0,001. Conclusão: A prevalência de fatores de risco cardiovascular em Natal – Brasil se constitui em grave problema de saúde pública, atingindo níveis que se igualam ou até superam os de outras cidades tanto no Brasil, como em outros países

Equações de segundo grau e mudança de variáveis

In this work are presented, as a review and in a historical context, the most used methods to solve quadratic equations. It is also shown the simplest type of change of variables, namely: x = Ay + B where A;B 2 R, and some changes of variables that were used to solve quadratic equations throughout history. Finally, a change of variable, which has been used by the author in the classroom as an alternative method, is presented and the result of this methodoly is illustrated by the responses of a test that was done by the students in classroom

A utilização da modelagem física no estudo do desenvolvimento, geometria e cinemática de falhas distensionais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior