Máquina de vetores-suporte intervalar

The Support Vector Machines (SVM) has attracted increasing attention in machine learning area, particularly on classification and patterns recognition. However, in some cases it is not easy to determinate accurately the class which given pattern belongs. This thesis involves the construction of a intervalar pattern classifier using SVM in association with intervalar theory, in order to model the separation of a pattern set between distinct classes with precision, aiming to obtain an optimized separation capable to treat imprecisions contained in the initial data and generated during the computational processing. The SVM is a linear machine. In order to allow it to solve real-world problems (usually nonlinear problems), it is necessary to treat the pattern set, know as input set, transforming from nonlinear nature to linear problem. The kernel machines are responsible to do this mapping. To create the intervalar extension of SVM, both for linear and nonlinear problems, it was necessary define intervalar kernel and the Mercer s theorem (which caracterize a kernel function) to intervalar function

Modelagem de máquina de indução trifásica sem mancais com bobinado dividido

This work presents a model of bearingless induction machine with divided winding. The main goal is to obtain a machine model to use a simpler control system as used in conventional induction machine and to know its behavior. The same strategies used in conventional machines were used to reach the bearingless induction machine model, which has made possible an easier treatment of the involved parameters. The studied machine is adapted from the conventional induction machine, the stator windings were divided and all terminals had been available. This method does not need an auxiliary stator winding for the radial position control which results in a more compact machine. Another issue about this machine is the variation of inductances array also present in result of the rotor displacement. The changeable air-gap produces variation in magnetic flux and in inductances consequently. The conventional machine model can be used for the bearingless machine when the rotor is centered, but in rotor displacement condition this model is not applicable. The bearingless machine has two sets of motor-bearing, both sets with four poles. It was constructed in horizontal position and this increases difficulty in implementation. The used rotor has peculiar characteristics; it is projected according to the stator to yield the greatest torque and force possible. It is important to observe that the current unbalance generated by the position control does not modify the machine characteristics, this only occurs due the radial rotor displacement. The obtained results validate the work; the data reached by a supervisory system corresponds the foreseen results of simulation which verify the model veracity

Proposta de uma máquina para colheita mecanizada de babaçu (Orbignya phalerata Mart.) para a agricultura familiar

O babaçu é uma planta de importância capital na economia de subsistência do norte do Brasil. Sua configuração sócio-ambiental o torna destaque na situação regional amazônica, onde os produtos advindos do babaçu possibilitam renda para a camada mais pobre da população amazônica, além da questão ambiental que é conotada à preservação dos babaçuais naturais. Um dos gargalos técnicos da produção do babaçu, em especial visando a extração do óleo de babaçu, é a colheita feita de forma manual e no sistema extrativista. O objetivo deste trabalho é propor o conceito de uma colhedora de babaçu moto-mecanizada, capaz de trabalhar em cultivos artificiais, assim como em florestas naturais. Foi utilizada a metodologia de projeto da matriz morfológica, onde foram elencadas as possíveis combinações de mecanismos e elementos para uma colhedora de babaçu. Como resultado foi obtido um conceito teórico, sendo concluída a viabilidade técnica de tal projeto, em estudos futuros pretende-se desenvolver estudos de viabilidade técnica detalhados, assim como estudos de viabilidade econômica.

Predição de promotores de Bacillus subtilis usando técnicas de aprendizado de máquina

One of the most important goals of bioinformatics is the ability to identify genes in uncharacterized DNA sequences on world wide database. Gene expression on prokaryotes initiates when the RNA-polymerase enzyme interacts with DNA regions called promoters. In these regions are located the main regulatory elements of the transcription process. Despite the improvement of in vitro techniques for molecular biology analysis, characterizing and identifying a great number of promoters on a genome is a complex task. Nevertheless, the main drawback is the absence of a large set of promoters to identify conserved patterns among the species. Hence, a in silico method to predict them on any species is a challenge. Improved promoter prediction methods can be one step towards developing more reliable ab initio gene prediction methods. In this work, we present an empirical comparison of Machine Learning (ML) techniques such as Na¨ýve Bayes, Decision Trees, Support Vector Machines and Neural Networks, Voted Perceptron, PART, k-NN and and ensemble approaches (Bagging and Boosting) to the task of predicting Bacillus subtilis. In order to do so, we first built two data set of promoter and nonpromoter sequences for B. subtilis and a hybrid one. In order to evaluate of ML methods a cross-validation procedure is applied. Good results were obtained with methods of ML like SVM and Naïve Bayes using B. subtilis. However, we have not reached good results on hybrid database

Aplicação de técnicas de aprendizado de máquina no reconhecimento de classes estruturais de proteínas

Nowadays, classifying proteins in structural classes, which concerns the inference of patterns in their 3D conformation, is one of the most important open problems in Molecular Biology. The main reason for this is that the function of a protein is intrinsically related to its spatial conformation. However, such conformations are very difficult to be obtained experimentally in laboratory. Thus, this problem has drawn the attention of many researchers in Bioinformatics. Considering the great difference between the number of protein sequences already known and the number of three-dimensional structures determined experimentally, the demand of automated techniques for structural classification of proteins is very high. In this context, computational tools, especially Machine Learning (ML) techniques, have become essential to deal with this problem. In this work, ML techniques are used in the recognition of protein structural classes: Decision Trees, k-Nearest Neighbor, Naive Bayes, Support Vector Machine and Neural Networks. These methods have been chosen because they represent different paradigms of learning and have been widely used in the Bioinfornmatics literature. Aiming to obtain an improvment in the performance of these techniques (individual classifiers), homogeneous (Bagging and Boosting) and heterogeneous (Voting, Stacking and StackingC) multiclassification systems are used. Moreover, since the protein database used in this work presents the problem of imbalanced classes, artificial techniques for class balance (Undersampling Random, Tomek Links, CNN, NCL and OSS) are used to minimize such a problem. In order to evaluate the ML methods, a cross-validation procedure is applied, where the accuracy of the classifiers is measured using the mean of classification error rate, on independent test sets. These means are compared, two by two, by the hypothesis test aiming to evaluate if there is, statistically, a significant difference between them. With respect to the results obtained with the individual classifiers, Support Vector Machine presented the best accuracy. In terms of the multi-classification systems (homogeneous and heterogeneous), they showed, in general, a superior or similar performance when compared to the one achieved by the individual classifiers used - especially Boosting with Decision Tree and the StackingC with Linear Regression as meta classifier. The Voting method, despite of its simplicity, has shown to be adequate for solving the problem presented in this work. The techniques for class balance, on the other hand, have not produced a significant improvement in the global classification error. Nevertheless, the use of such techniques did improve the classification error for the minority class. In this context, the NCL technique has shown to be more appropriated

Estudo avaliativo de um algoritmo genético auto-organizável e multiobjetivo utilizando aprendizado de máquina para aplicações de telecomunicações

This paper presents an evaluative study about the effects of using a machine learning technique on the main features of a self-organizing and multiobjective genetic algorithm (GA). A typical GA can be seen as a search technique which is usually applied in problems involving no polynomial complexity. Originally, these algorithms were designed to create methods that seek acceptable solutions to problems where the global optimum is inaccessible or difficult to obtain. At first, the GAs considered only one evaluation function and a single objective optimization. Today, however, implementations that consider several optimization objectives simultaneously (multiobjective algorithms) are common, besides allowing the change of many components of the algorithm dynamically (self-organizing algorithms). At the same time, they are also common combinations of GAs with machine learning techniques to improve some of its characteristics of performance and use. In this work, a GA with a machine learning technique was analyzed and applied in a antenna design. We used a variant of bicubic interpolation technique, called 2D Spline, as machine learning technique to estimate the behavior of a dynamic fitness function, based on the knowledge obtained from a set of laboratory experiments. This fitness function is also called evaluation function and, it is responsible for determining the fitness degree of a candidate solution (individual), in relation to others in the same population. The algorithm can be applied in many areas, including in the field of telecommunications, as projects of antennas and frequency selective surfaces. In this particular work, the presented algorithm was developed to optimize the design of a microstrip antenna, usually used in wireless communication systems for application in Ultra-Wideband (UWB). The algorithm allowed the optimization of two variables of geometry antenna - the length (Ls) and width (Ws) a slit in the ground plane with respect to three objectives: radiated signal bandwidth, return loss and central frequency deviation. These two dimensions (Ws and Ls) are used as variables in three different interpolation functions, one Spline for each optimization objective, to compose a multiobjective and aggregate fitness function. The final result proposed by the algorithm was compared with the simulation program result and the measured result of a physical prototype of the antenna built in the laboratory. In the present study, the algorithm was analyzed with respect to their success degree in relation to four important characteristics of a self-organizing multiobjective GA: performance, flexibility, scalability and accuracy. At the end of the study, it was observed a time increase in algorithm execution in comparison to a common GA, due to the time required for the machine learning process. On the plus side, we notice a sensitive gain with respect to flexibility and accuracy of results, and a prosperous path that indicates directions to the algorithm to allow the optimization problems with "η" variables

Controlador adaptativo por posicionamento de pólos e estrutura variável aplicado ao controle de geradores eólicos baseados em máquina de indução duplamente alimentada

In the last decade, the renewable energy sources have present a major propulsion in the world due to several factors: political, environmental, financial and others. Within this context, we have in particular the energy obtained through wind, wind energy - that has highlighted with rapid growth in recent years, including in Brazil, mostly in the Northeast, due to it s benefit-cost between the clean energies. In this context, we propose to compare the variable structure adaptive pole placement control (VS-APPC) with a traditional control technique proportional integral controller (PI), applied to set the control of machine side in a conversion system using a wind generator based on Double-Fed Induction Generator (DFIG). Robustness and performance tests were carried out to the uncertainties of the internal parameters of the machine and variations of speed reference.

Uma máquina de redução de grafos para serviços web

Web services are software accessible via the Internet that provide functionality to be used by applications. Today, it is natural to reuse third-party services to compose new services. This process of composition can occur in two styles, called orchestration and choreography. A choreography represents a collaboration between services which know their partners in the composition, to achieve the service s desired functionality. On the other hand, an orchestration have a central process (the orchestrator) that coordinates all application operations. Our work is placed in this latter context, by proposing an abstract model for running service orchestrations. For this purpose, a graph reduction machine will be defined for the implementation of service orchestrations specified in a variant of the PEWS composition language. Moreover, a prototype of this machine (in Java) is built as a proof of concept

Detecção e recuperação de falhas para a máquina de redução de grafos PEWS-AM

Web services are software units that allow access to one or more resources, supporting the deployment of business processes in the Web. They use well-defined interfaces, using web standard protocols, making possible the communication between entities implemented on different platforms. Due to these features, Web services can be integrated as services compositions to form more robust loose coupling applications. Web services are subject to failures, unwanted situations that may compromise the business process partially or completely. Failures can occur both in the design of compositions as in the execution of compositions. As a result, it is essential to create mechanisms to make the implementation of service compositions more robust and to treat failures. Specifically, we propose the support for fault recovery in service compositions described in PEWS language and executed on PEWS-AM, an graph reduction machine. To support recovery failure on PEWS-AM, we extend the PEWS language specification and adapted the rules of translation and reduction of graphs for this machine. These contributions were made both in the model of abstract machine as at the implementation level

Recurso de interface captcha: segurança e usabilidade na distinção entre humano e máquina

Com o intuito de impedir que computadores enviem mensagens automáticas se passando por usuários reais, desenvolvedores tem utilizado o recurso de interface CAPTCHA para distinguir o preenchimento de dados e submissões realizadas por humanos e por máquinas. Este trabalho apresenta as principais modalidades de CAPTCHAs e discute as implicações na usabilidade. Para tanto, foram aplicados questionários e realizados testes de usabilidade em três modalidades de CAPTCHA. Verificou-se que, embora os usuários tenham consciência da importância do uso do CAPTCHA como ferramenta de segurança, percebeu-se que tal recurso compromete a usabilidade, gerando insatisfação e em alguns casos, a desistência da realização da tarefa

Modelagem de máquina pneumática recolhedora de frutos de café em terreiro usando análise por elementos finitos

Teve-se o objetivo de desenvolver um modelo matemático por meio de análise de elementos finitos, utilizando o programa computacional ANSYS®, versão 5.7, para otimizar o projeto de máquina recolhedora de frutos de café no terreiro. A modelagem da máquina foi realizada com base no levantamento das características aerodinâmicas dos frutos de café e da vazão de ar necessária para o transporte pneumático dos frutos. Foram obtidas, experimentalmente, as pressões estáticas nos dutos da máquina, sendo esses valores comparados com os resultados determinados pelo programa ANSYS, no intuito de validar o modelo. Com base nos resultados numéricos obtidos, concluiu-se que a modelagem desenvolvida apresentou resultados próximos aos determinados experimentalmente, obtendo erro relativo médio nos valores simulados de pressão de 9,2%. Por meio da modelagem, identificaram-se faixas de pressão que dificultariam o transporte pneumático dos frutos de café em alguns pontos da máquina. Esses problemas foram corrigidos e, com isso, o fluxo de ar proporcionado pelo ventilador foi suficiente para succionar os frutos de café no terreiro e transportá-los para dentro do reservatório da máquina. A modelagem desenvolvida atendeu às necessidades propostas no trabalho para o recolhimento dos frutos de café utilizando transporte pneumático eficientemente.

Aprendizado de máquina simbólico e técnicas fractais para caracterizar rejeição em biópsia miocárdica

This work combines symbolic machine learning and multiscale fractal techniques to generate models that characterize cellular rejection in myocardial biopsies and that can base a diagnosis support system. The models express the knowledge by the features threshold, fractal dimension, lacunarity, number of clusters, spatial percolation and percolation probability, all obtained with myocardial biopsies processing. Models were evaluated and the most significant was the one generated by the C4.5 algorithm for the features spatial percolation and number of clusters. The result is relevant and contributes to the specialized literature since it determines a standard diagnosis protocol. © 2013 Springer.

Um novo sistema de refrigeração com controle de temperatura, compressor aberto, máquina de indução trifásica com velocidade variável e correção ativa do fator de potência do estágio de entrada

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)