Estudo de associação genômica e análise de enriquecimento genético da criotolerância espermática em bovinos da raça Nelore

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Cobertura populacional do Sistema de Vigilância Alimentar e Nutricional no Estado de São Paulo, Brasil

O objetivo deste artigo é estimar a cobertura populacional do Sistema de Vigilância Alimentar e Nutricional (SISVAN) nos diferentes estágios de vida e avaliar seu funcionamento no estado de São Paulo. O estudo incluiu 65 municípios divididos em 14 regiões do estado. A cobertura do SISVAN foi estimada a partir de dados de monitoramento do estado nutricional disponíveis nos relatórios públicos, e do número de usuários que frequentam os serviços públicos de saúde. O total de usuários foi obtido pela diferença entre o total de habitantes e o número de beneficiários de planos de saúde privados. A maioria das regiões apresentou uma cobertura reduzida (<10%). Cerca de 57% revelaram cobertura entre 5 e 10%. Constatou-se uma preponderância de registros do estado nutricional de crianças para todas as regiões do Estado. Chama a atenção a reduzida cobertura entre os idosos, que é inexistente ou próxima de zero na maioria das regiões. Apesar dos esforços empreendidos pelo governo visando à ampliação e à qualificação do SISVAN, o monitoramento nutricional no estado de São Paulo ainda é insuficiente. Esta condição compromete sua utilização na elaboração de políticas efetivas na área de alimentação e nutrição.

Análise de modelos troposféricos no posicionamento baseado em redes usando o conceito de VRS

In the past few years several GPS (Global Position System) positioning techniques have been develope and/or improved with the goal of obtaining high accuracy and productivity in real time. The reference station network concept besides to enabling quality and reliability in positioning for scientific and civil GPS community, allows studies concerning tropospheric refraction modeling in the network region. Moreover, among the network corrections transmission methods available to users, there is the VRS (Virtual Reference Station) concept. In this method, the data of a virtual station are generated near the rover receiver (user). This provides a short baseline and the user has the possibility of using a single frequency receiver to accomplish the relative positioning. In this paper, the methodology applied to generate VRS data, using different tropospheric models is described. Thus, comparative tests were conducted in the four seasons with the NWP/INPE (Numerical Weather Prediction/National Institute for Space Research) and Hopfield tropospheric models. In order to analyse the VRS data quality, it was used the Precise Point Positioning (PPP) method, where satisfactory results were found. Mean differences between PNT/INPE and Hopfield models of 9.75% and 24.2% for the hydrostatic and wet days, respectively were obtained.

Análise de séries temporais de multicaminho em estações de monitoramento contínuo

Data from reference stations are widely used in GNSS (Global Navigation Satellite System) positioning, and can be used in relative positioning or network-based positioning concept. Positioning accuracy will be directly influenced by errors in signals collected in these stations. In this paper, it is aimed at evaluating these data quality using temporal series of multipath index MP1 and MP2. A statistical study of temporal series with 7 years of daily observations related to 7 stations from RBMC (Rede Brasileira de Monitoramento Contínuo) was accomplished. In order to investigate trends and seasonality a linear regression model, correlograms, and Fourier periodograms were used. We also used a harmonic adjust to identify peaks on temporal series. At last, the possible causes of seasonality found in some stations were discussed. It was also possible to identify peaks in MP values of March and October months (mainly in stations located near geomagnetic equator).

A path- and label-cost propagation approach to speedup the training of the optimum-path forest classifier

In general, pattern recognition techniques require a high computational burden for learning the discriminating functions that are responsible to separate samples from distinct classes. As such, there are several studies that make effort to employ machine learning algorithms in the context of big data classification problems. The research on this area ranges from Graphics Processing Units-based implementations to mathematical optimizations, being the main drawback of the former approaches to be dependent on the graphic video card. Here, we propose an architecture-independent optimization approach for the optimum-path forest (OPF) classifier, that is designed using a theoretical formulation that relates the minimum spanning tree with the minimum spanning forest generated by the OPF over the training dataset. The experiments have shown that the approach proposed can be faster than the traditional one in five public datasets, being also as accurate as the original OPF. (C) 2014 Elsevier B. V. All rights reserved.

Desempenho de bovinos cruzados e parâmetros qualitativos de músculos maturados

Pós-graduação em Zootecnia - FCAV

Análise do impacto de sistemas fotovoltaicos conectados à rede de distribuição na qualidade da energia de uma smart city: Análise do impacto de sistemas fotovoltaicos conectados à rede de distribuição na qualidade da energia de uma smart city

Pós-graduação em Engenharia Elétrica - FEIS