Utilização de redes neurais artificiais aplicadas na discriminação de padrões de doenças florestais

Pós-graduação em Ciência Florestal - FCA

Uso de Redes Neurais ARTMAP Nebulosas para a classificação de padrões em sinais ECoG relacionados ao movimento dos dedos

Pós-graduação em Engenharia Elétrica - FEIS

Combinando sistemas de informação geográfica e redes neurais GMDH para prever densidades demográficas

The aim of this work is to advance a new approach for estimating demographic density, through combining a Geographic Information System with GMDH Neural Networks. The model that is suggested parts the analyzed space into a rectangular grid formed by multiple cells measuring 0.01 km2 each. The forecasts are elaborated based on the demographic density in each cell and in its neighboring cells at a given time. Despite the limited availability of data during the modeling phase, the utilization of this method for studying a Brazilian medium-sized city presented promising results.

Obtenção de neurônios de redes neurais de base radial via agrupamento de dados por floresta de caminhos ótimos

Neste artigo é apresentada uma abordagem para aumentar a eficácia das Redes Neurais Artificiais de Funções de Base Radial utilizando um algoritmo de agrupamento de dados via Floresta de Caminhos Ótimos. Algumas técnicas comumente empregadas para essa tarefa, como o conhecido k-médias, requerem um determinado número de classes/agrupamentos prévio à sua execução. Embora o número de classes seja conhecido em problemas supervisionados, o número real de agrupamentos é difícil de ser encontrado, dado que uma classe pode ser representada por mais de um agrupamento. Experimentos em nove bases de dados, em conjunto com análises estatísticas, demonstraram que o classificador por Floresta de Caminhos Ótimos possui um melhor desempenho que a técnica k-médias, bem como encontra as médias das distribuições Gaussianas em posições muito similares às encontradas por este último. Entretanto, o classificador por Floresta de Caminhos Ótimos possui um custo computacional maior, dado que a sua etapa de treinamento é mais custosa que a da técnica k-médias.

Utilização dos parâmetros físico-químicos e redes neurais artificiais na identificação dos méis de abelhas (Apis mellifera L.) produzidos no verão e inverno na microrregião de Campos de Jordão, São Paulo

The objective of this work was to typify, through physicochemical parameters, honey from Campos do Jordão’s microrregion, and verify how samples are grouped in accordance with the climatic production seasonality (summer and winter). It were assessed 30 samples of honey from beekeepers located in the cities of Monteiro Lobato, Campos do Jordão, Santo Antonio do Pinhal e São Bento do Sapucaí-SP, regarding both periods of honey production (November to February; July to September, during 2007 and 2008; n = 30). Samples were submitted to physicochemical analysis of total acidity, pH, humidity, water activity, density, aminoacids, ashes, color and electrical conductivity, identifying physicochemical standards of honey samples from both periods of production. Next, we carried out a cluster analysis of data using k-means algorithm, which grouped the samples into two classes (summer and winter). Thus, there was a supervised training of an Artificial Neural Network (ANN) using backpropagation algorithm. According to the analysis, the knowledge gained through the ANN classified the samples with 80% accuracy. It was observed that the ANNs have proved an effective tool to group samples of honey of the region of Campos do Jordao according to their physicochemical characteristics, depending on the different production periods.

Mapeamento da cinemática inversa de um manipulador robótico utilizando redes neurais artificiais configuradas em paralelo

Pós-graduação em Engenharia Elétrica - FEIS

Redes neurais artificiais na previsão da inflação: aplicação como ferramenta de apoio à análise de decisões financeiras em organizações de pequeno porte

As estimações das taxas de inflação são de fundamental importância para os gestores, pois as decisões de investimento estão intimamente ligadas a elas. Contudo, o comportamento inflacionário tende a ser não linear e até mesmo caótico, tornando difícil a sua correta estimação. Essa característica do fenômeno pode tornar imprecisos os modelos mais simples de previsão, acessíveis às pequenas organizações, uma vez que muitos deles necessitam de grandes manipulações de dados e/ou softwares especializados. O presente artigo tem por objetivo avaliar, por meio de análise formal estatística, a eficácia das redes neurais artificiais (RNA) na previsão da inflação, dentro da realidade de organizações de pequeno porte. As RNA são ferramentas adequadas para mensurar os fenômenos inflacionários, por se tratar de aproximações de funções polinomiais, capazes de lidar com fenômenos não lineares. Para esse processo, foram selecionados três modelos básicos de redes neurais artificiais Multi Layer Perceptron, passíveis de implementação a partir de planilhas eletrônicas de código aberto. Os três modelos foram testados a partir de um conjunto de variáveis independentes sugeridas por Bresser-Pereira e Nakano (1984), com defasagem de um, seis e doze meses. Para tal, foram utilizados testes de Wilcoxon, coeficiente de determinação R² e o percentual de erro médio dos modelos. O conjunto de dados foi dividido em dois, sendo um grupo usado para treinamento das redes neurais artificiais, enquanto outro grupo era utilizado para verificar a capacidade de predição dos modelos e sua capacidade de generalização. Com isso, o trabalho concluiu que determinados modelos de redes neurais artificiais têm uma razoável capacidade de predição da inflação no curto prazo e se constituem em uma alternativa razoável para esse tipo de mensuração.

Contexto e modularização em redes neurais recorrentes para aprendizagem de seqüências temporais

Este trabalho apresenta um sistema neural modular, que processa separadamente informações de contexto espacial e temporal, para a tarefa de reprodução de sequências temporais. Para o desenvolvimento do sistema neural foram considerados redes neurais recorrentes, modelos estocásticos, sistemas neurais modulares e processamento de informações de contexto. Em seguida, foram estudados três modelos com abordagens distintas para aprendizagem de seqüências temporais: uma rede neural parcialmente recorrente, um exemplo de sistema neural modular e um modelo estocástico utilizando a teoria de modelos markovianos escondidos. Com base nos estudos e modelos apresentados, esta pesquisa propõe um sistema formado por dois módulos sucessivos distintos. Uma rede de propagação direta (módulo estimador de contexto espacial) realiza o processamento de contexto espacial identificando a seqüência a ser reproduzida e fornecendo um protótipo do contexto para o segundo módulo. Este é formado por uma rede parcialmente recorrente (módulo de reprodução de sequências temporais) para aprender as informações de contexto temporal e reproduzir em suas saídas a seqüência identificada pelo módulo anterior. Para a finalidade mencionada, este mestrado utiliza a distribuição de Gibbs na saída do módulo para contexto espacial de forma que este forneça probabilidades de contexto espacial, indicando o grau de certeza do módulo e possibilitando a utilização de procedimentos especiais para os casos de dúvida. O sistema neural foi testado em conjuntos contendo trajetórias abertas, fechadas, e com diferentes situações de ambigüidade e complexidade. Duas situações distintas foram avaliadas: (a) capacidade do sistema em reproduzir trajetórias a partir de pontos iniciais treinados; e (b) capacidade de generalização do sistema reproduzindo trajetórias considerando pontos iniciais ou finais em situações não treinadas. A situação (b) é um problema de difícil ) solução em redes neurais devido à falta de contexto temporal, essencial na reprodução de seqüências. Foram realizados experimentos comparando o desempenho do sistema modular proposto com o de uma rede parcialmente recorrente operando sozinha e um sistema modular neural (TOTEM). Os resultados sugerem que o sistema proposto apresentou uma capacidade de generalização significamente melhor, sem que houvesse uma deterioração na capacidade de reproduzir seqüências treinadas. Esses resultados foram obtidos em sistema mais simples que o TOTEM.

Cálculo das incertezas climáticas sobre América do Sul utilizando modelos do CMIP5: aferição através das redes neurais artificiais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Avaliação da aplicabilidade de correlações matemáticas e redes neurais na predição de parâmetros de especificação do diesel

Diesel fuel is one of leading petroleum products marketed in Brazil, and has its quality monitored by specialized laboratories linked to the National Agency of Petroleum, Natural Gas and Biofuels - ANP. The main trial evaluating physicochemical properties of diesel are listed in the resolutions ANP Nº 65 of December 9th, 2011 and Nº 45 of December 20th, 2012 that determine the specification limits for each parameter and methodologies of analysis that should be adopted. However the methods used although quite consolidated, require dedicated equipment with high cost of acquisition and maintenance, as well as technical expertise for completion of these trials. Studies for development of more rapid alternative methods and lower cost have been the focus of many researchers. In this same perspective, this work conducted an assessment of the applicability of existing specialized literature on mathematical equations and artificial neural networks (ANN) for the determination of parameters of specification diesel fuel. 162 samples of diesel with a maximum sulfur content of 50, 500 and 1800 ppm, which were analyzed in a specialized laboratory using ASTM methods recommended by the ANP, with a total of 810 trials were used for this study. Experimental results atmospheric distillation (ASTM D86), and density (ASTM D4052) of diesel samples were used as basic input variables to the equations evaluated. The RNAs were applied to predict the flash point, cetane number and sulfur content (S50, S500, S1800), in which were tested network architectures feed-forward backpropagation and generalized regression varying the parameters of the matrix input in order to determine the set of variables and the best type of network for the prediction of variables of interest. The results obtained by the equations and RNAs were compared with experimental results using the nonparametric Wilcoxon test and Student's t test, at a significance level of 5%, as well as the coefficient of determination and percentage error, an error which was obtained 27, 61% for the flash point using a specific equation. The cetane number was obtained by three equations, and both showed good correlation coefficients, especially equation based on aniline point, with the lowest error of 0,816%. ANNs for predicting the flash point and the index cetane showed quite superior results to those observed with the mathematical equations, respectively, with errors of 2,55% and 0,23%. Among the samples with different sulfur contents, the RNAs were better able to predict the S1800 with error of 1,557%. Generally, networks of the type feedforward proved superior to generalized regression.

Implementação de uma matriz de neurônios dinamicamente reconfigurável para descrição de topologias de redes neurais artificiais multilayer perceptrons

The Artificial Neural Networks (ANN), which is one of the branches of Artificial Intelligence (AI), are being employed as a solution to many complex problems existing in several areas. To solve these problems, it is essential that its implementation is done in hardware. Among the strategies to be adopted and met during the design phase and implementation of RNAs in hardware, connections between neurons are the ones that need more attention. Recently, are RNAs implemented both in application specific integrated circuits's (Application Specific Integrated Circuits - ASIC) and in integrated circuits configured by the user, like the Field Programmable Gate Array (FPGA), which have the ability to be partially rewritten, at runtime, forming thus a system Partially Reconfigurable (SPR), the use of which provides several advantages, such as flexibility in implementation and cost reduction. It has been noted a considerable increase in the use of FPGAs for implementing ANNs. Given the above, it is proposed to implement an array of reconfigurable neurons for topologies Description of artificial neural network multilayer perceptrons (MLPs) in FPGA, in order to encourage feedback and reuse of neural processors (perceptrons) used in the same area of the circuit. It is further proposed, a communication network capable of performing the reuse of artificial neurons. The architecture of the proposed system will configure various topologies MLPs networks through partial reconfiguration of the FPGA. To allow this flexibility RNAs settings, a set of digital components (datapath), and a controller were developed to execute instructions that define each topology for MLP neural network.

Implementação de uma matriz de neurônios dinamicamente reconfigurável para descrição de topologias de redes neurais artificiais multilayer perceptrons

The Artificial Neural Networks (ANN), which is one of the branches of Artificial Intelligence (AI), are being employed as a solution to many complex problems existing in several areas. To solve these problems, it is essential that its implementation is done in hardware. Among the strategies to be adopted and met during the design phase and implementation of RNAs in hardware, connections between neurons are the ones that need more attention. Recently, are RNAs implemented both in application specific integrated circuits's (Application Specific Integrated Circuits - ASIC) and in integrated circuits configured by the user, like the Field Programmable Gate Array (FPGA), which have the ability to be partially rewritten, at runtime, forming thus a system Partially Reconfigurable (SPR), the use of which provides several advantages, such as flexibility in implementation and cost reduction. It has been noted a considerable increase in the use of FPGAs for implementing ANNs. Given the above, it is proposed to implement an array of reconfigurable neurons for topologies Description of artificial neural network multilayer perceptrons (MLPs) in FPGA, in order to encourage feedback and reuse of neural processors (perceptrons) used in the same area of the circuit. It is further proposed, a communication network capable of performing the reuse of artificial neurons. The architecture of the proposed system will configure various topologies MLPs networks through partial reconfiguration of the FPGA. To allow this flexibility RNAs settings, a set of digital components (datapath), and a controller were developed to execute instructions that define each topology for MLP neural network.

Classificação de nódulos pulmonares em imagens tomográficas utilizando redes neurais artificiais em cascata

Lung cancer is the most common of malignant tumors, with 1.59 million new cases worldwide in 2012. Early detection is the main factor to determine the survival of patients affected by this disease. Furthermore, the correct classification is important to define the most appropriate therapeutic approach as well as suggest the prognosis and the clinical disease evolution. Among the exams used to detect lung cancer, computed tomography have been the most indicated. However, CT images are naturally complex and even experts medical are subject to fault detection or classification. In order to assist the detection of malignant tumors, computer-aided diagnosis systems have been developed to aid reduce the amount of false positives biopsies. In this work it was developed an automatic classification system of pulmonary nodules on CT images by using Artificial Neural Networks. Morphological, texture and intensity attributes were extracted from lung nodules cut tomographic images using elliptical regions of interest that they were subsequently segmented by Otsu method. These features were selected through statistical tests that compare populations (T test of Student and U test of Mann-Whitney); from which it originated a ranking. The features after selected, were inserted in Artificial Neural Networks (backpropagation) to compose two types of classification; one to classify nodules in malignant and benign (network 1); and another to classify two types of malignancies (network 2); featuring a cascade classifier. The best networks were associated and its performance was measured by the area under the ROC curve, where the network 1 and network 2 achieved performance equal to 0.901 and 0.892 respectively.

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