925 resultados para probabilistic principal component analysis (probabilistic PCA)


Relevância:

100.00% 100.00%

Publicador:

Resumo:

In this paper we deal with the problem of feature selection by introducing a new approach based on Gravitational Search Algorithm (GSA). The proposed algorithm combines the optimization behavior of GSA together with the speed of Optimum-Path Forest (OPF) classifier in order to provide a fast and accurate framework for feature selection. Experiments on datasets obtained from a wide range of applications, such as vowel recognition, image classification and fraud detection in power distribution systems are conducted in order to asses the robustness of the proposed technique against Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and a Particle Swarm Optimization (PSO)-based algorithm for feature selection.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

This paper aims to study the ecological system of the Pardo River, at the source and lower-order passages, which are in the Botucatu area, São Paulo State, Brazil. This study was carried out to determine water quality with some chemical-physical indicators, coliforms, and chemical species of samples taken monthly, 1995/02-1996/01, from eight sampling stations sited along the Pardo River. The results in the river monitoring are discussed based on annual averages, analysis of variance, and compared to Tukey's Studentized Range-HSD, and principal component analysis (PCA) was applied to normalize data to assess association between variables. We can conclude that the variables used are very efficient for identifying and that the dry season shows the worst water quality. These were caused by organic matter, nutrients (originate) from anthropogenic sources (spatial sources) and mainly municipal wastewater, affecting the quality and hydrochemistry of the river water, which have been differentiated and assigned to polluting sources. Meanwhile, the degree of degradation of the Pardo River is low (sewage treatment carried out by the city of Pardinho is efficient), leaving the water of the river suitable for use by the population of Botucatu, after conventional treatment (Conama, Resolucao No. 20, CONAMA, Brazilia DF, 09-23, 1986-the water of the Pardo river is classified as level 03). (C) 2001 Elsevier B.V. Ltd. All rights reserved.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

A number of attempts have been made to obtain a clear definition of biological stress. However, in spite of the efforts, some controversies on the concept of plant stress remain. The current versions are centered either on the cause (stress factor) or on the effect (stress response) of environmental stress. The objective of this study was to contribute to the definition of stress, using a hierarchical approach. Thus, we have performed an analysis of the most usual stress concepts and tested the relevance of considering different observation scales in a study on plant response to water deficit. Seedlings of Eucalyptus grandis were grown in vitro at water potentials ranging from -0.16 to -0.6 MPa, and evaluated according to growth and biochemical parameters. Data were analyzed through principal component analysis (PCA), which pointed to a hierarchical organization in plant responses to environmental disturbances. Growth parameters (height and dry weight) are more sensitive to water deficit than biochemical ones (sugars, proline, and protein), suggesting that higher hierarchical levels were more sensitive to environmental constraints than lower hierarchical ones. We suggest that before considering an environmental fluctuation as stressful, it is necessary to take into account different levels of plant response, and that the evaluation of the effects of environmental disturbances on an organism depends on the observation scale being used. Hence, a more appropriate stress concept should consider the hierarchical organization of the biological systems, not only for a more adequate theoretical approach, but also for the improvement of practical studies on plants under stress.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

As raças e populações de pupunheiras ao longo dos rios Amazonas e Solimões apresentam grande variabilidade genética ainda não totalmente caracterizada. Neste estudo, foram aplicadas técnicas estatísticas multivariadas a 15 descritores morfológicos numa tentativa de caracterizar, morfometricamente, três raças existentes ao longo da Bacia desses rios. As três análises em conjunto permitiram uma discriminação das raças, mostrando também que os descritores mais importantes nessa seleção foram: número de espigas, comprimento da ráquis, peso do fruto, espessura das cascas, facilidade para descascar os frutos, peso das cascas, sabor dos frutos, espessura da polpa, distância morfológica dos frutos e peso da semente.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Os lagos de pesca recreativa, pesque-pague, surgiram no Brasil em um cenário agrícola denominado como novo rural brasileiro. Este estudo conduzido no noroeste paulista teve como foco o desempenho produtivo dos lagos de pesca recreativa. Foram feitas visitas mensais a nove empreendimentos de pesca recreativa durante seis meses. A cada visita foi aplicado um questionário contendo 13 indicadores de desempenho. Os dados levantados foram submetidos à análise multivariada (MANOVA), análise de componente principal (ACP) e análise de agrupamento. A MANOVA indicou diferenças significativas entre os lagos de pesca recreativa. A ACP revelou, a partir do coeficiente dos autovalores, três atributos: sistema produtivo, gerenciamento pesqueiro e administração operacional. A análise de agrupamento classificou os lagos de pesca recreativa em quatro grupos. Freqüência de pescadores (FP), densidade de estocagem (DE), biomassa de estocagem (BE), captura total (CT) e captura lago dia (CLD), os quais fazem parte do atributo sistema produtivo, mostrando-se os indicadores mais importantes para a avaliação de desempenho dos lagos de pesca recreativa neste estudo.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Relevância:

100.00% 100.00%

Publicador:

Resumo:

In conformational analysis, the systematic search method completely maps the space but suffers from the combinatorial explosion problem because the number of conformations increases exponentially with the number of free rotation angles. This study introduces a new methodology of conformational analysis that controls the combinatorial explosion. It is based on a dimensional reduction of the system through the use of principal component analysis. The results are exactly the same as those obtained for the complete search but, in this case, the number of conformations increases only quadratically with the number of free rotation angles. The method is applied to a series of three drugs: omeprazole. pantoprazole, lansoprazole-benzimidazoles that suppress gastric-acid secretion by means of H(+), K(+)-ATPase enzyme inhibition. (C) 2002 John Wiley Sons. Inc.