Particle Competition and Cooperation in Networks for Semi-Supervised Learning

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

Structural study of thin films prepared from tungstate glass matrix by Raman and X-ray absorption spectroscopy

Thin films were prepared using glass precursors obtained in the ternary system NaPO(3)-BaF(2)-WO(3) and the binary system NaPO(3)-WO(3) with high concentrations of WO(3) (above 40% molar). Vitreous samples have been used as a target to prepare thin films. Such films were deposited using the electron beam evaporation method onto soda-lime glass substrates. Several structural characterizations were performed by Raman spectroscopy and X-ray Absorption Near Edge Spectroscopy (XANES) at the tungsten L(I) and L(III) absorption edges. XANES investigations showed that tungsten atoms are only sixfold coordinated (octahedral WO(6)) and that these films are free of tungstate tetrahedral units (WO(4)). In addition, Raman spectroscopy allowed identifying a break in the linear phosphate chains as the amount of WO(3) increases and the formation of P-O-W bonds in the films network indicating the intermediary behavior of WO(6) octahedra in the film network. Based on XANES data, we suggested a new attribution of several Raman absorption bands which allowed identifying the presence of W-O and W=O terminal bonds and a progressive apparition of W-O-W bridging bonds for the most WO(3) concentrated samples (above 40% molar) attributed to the formation of WO(6) clusters. (C) 2008 Elsevier B.V. All rights reserved.

As possibilidades de uso de ferramentas da informática: software para análise de dados qualitativos e ISD

The objective of the present article is to identify and discuss the possibilities of using qualitative data analysis software in the framework of procedures proposed by SDI (socio-discursive interactionism), emphasizing free distribuited software or free versions of commercial software. A literature review of software for qualitative data analysis in the area of social sciences and humanities, focusing on language studies is presented. Some tools, such as: Wef-tQDA, MLCT, Yoshikoder and Tropes are examined with their respective features and functions. The software called Tropes is examined in more detail because of its particular relation with language and semantic analysis, as well as its embeded classification of linguistic elements such as, types of verbs, adjectives, modalizations, etc. Although trying to completely automate an SDI based analysis is not feasible, the programs appear to be powerful helpers in analyzing specific questions. Still, it seems important to be familiar with software options and use different applications in order to obtain a more diversified vision of the data. It is up to the researcher to be critical of the analysis provided by the machine.

Structural study of tungstate fluorophosphate glasses by Raman and X-ray absorption spectroscopy

Transparent glasses were synthesized in the NaPO3-BaF2 WO3 tertiary system and several structural characterizations were performed by X-ray absorption spectroscopy (XANES) at the tungsten L-I and L-III absorption edges and by Raman spectroscopy. Special attention was paid to the coordination state of tungsten atoms in the vitreous network.XANES investigations showed that tungsten atoms are only six-fold coordinated (octahedra WO6) and that these glasses are free of tungstate tetrahedra (WO4).In addition, Raman spectroscopy allowed to identify a break in the linear phosphate chains as the amount of WO3 increases and the formation of P-O-W bonds in the vitreous network indicating the modifier behavior of WO6 octahedra in the glass network. Based on XANES data, we suggested a new attribution of several Raman absorption bands which allowed to identify the presence of W-O- and W=O terminal bonds and a progressive apparition of W-O-W bridging bonds for the most WO3 concentrated samples (≥ 30% molar) due to the formation of WO6 clusters. © 2004 Elsevier B.V. All rights reserved.

Application of ARX neural networks to model the rate of penetration of petroleum wells drilling

Bit performance prediction has been a challenging problem for the petroleum industry. It is essential in cost reduction associated with well planning and drilling performance prediction, especially when rigs leasing rates tend to follow the projects-demand and barrel-price rises. A methodology to model and predict one of the drilling bit performance evaluator, the Rate of Penetration (ROP), is presented herein. As the parameters affecting the ROP are complex and their relationship not easily modeled, the application of a Neural Network is suggested. In the present work, a dynamic neural network, based on the Auto-Regressive with Extra Input Signals model, or ARX model, is used to approach the ROP modeling problem. The network was applied to a real oil offshore field data set, consisted of information from seven wells drilled with an equal-diameter bit.

Geração de dados GPS de pseudodistância para uma estação virtual: Métodos, implementação e análise dos resultados

This paper aims to evaluate the quality of the pseudorange observables generated for a Virtual Reference Station (VRS). In order to generate the VRS data three different approaches were implemented and tested. In the first one, raw data from the reference station network were used while in the second it was based on double difference reference station corrections. Finally, in the third approach atmospheric models (ionosphere and troposphere) were used to create the VRS data. Sao Paulo State Network stations were used in all experiments. The VRS data were generated in a reference station position of known coordinates (real file). In order to validate the approaches, the VRS data were compared with the real data file. The results were quite similar, reaching the decimeter or centimeter level, depending on the approach applied.

Particle competition and cooperation to prevent error propagation from mislabeled data in semi-supervised learning

Semi-supervised learning is applied to classification problems where only a small portion of the data items is labeled. In these cases, the reliability of the labels is a crucial factor, because mislabeled items may propagate wrong labels to a large portion or even the entire data set. This paper aims to address this problem by presenting a graph-based (network-based) semi-supervised learning method, specifically designed to handle data sets with mislabeled samples. The method uses teams of walking particles, with competitive and cooperative behavior, for label propagation in the network constructed from the input data set. The proposed model is nature-inspired and it incorporates some features to make it robust to a considerable amount of mislabeled data items. Computer simulations show the performance of the method in the presence of different percentage of mislabeled data, in networks of different sizes and average node degree. Importantly, these simulations reveals the existence of the critical points of the mislabeled subset size, below which the network is free of wrong label contamination, but above which the mislabeled samples start to propagate their labels to the rest of the network. Moreover, numerical comparisons have been made among the proposed method and other representative graph-based semi-supervised learning methods using both artificial and real-world data sets. Interestingly, the proposed method has increasing better performance than the others as the percentage of mislabeled samples is getting larger. © 2012 IEEE.

Computer techniques towards the automatic characterization of graphite particles in metallographic images of industrial materials

The automatic characterization of particles in metallographic images has been paramount, mainly because of the importance of quantifying such microstructures in order to assess the mechanical properties of materials common used in industry. This automated characterization may avoid problems related with fatigue and possible measurement errors. In this paper, computer techniques are used and assessed towards the accomplishment of this crucial industrial goal in an efficient and robust manner. Hence, the use of the most actively pursued machine learning classification techniques. In particularity, Support Vector Machine, Bayesian and Optimum-Path Forest based classifiers, and also the Otsu's method, which is commonly used in computer imaging to binarize automatically simply images and used here to demonstrated the need for more complex methods, are evaluated in the characterization of graphite particles in metallographic images. The statistical based analysis performed confirmed that these computer techniques are efficient solutions to accomplish the aimed characterization. Additionally, the Optimum-Path Forest based classifier demonstrated an overall superior performance, both in terms of accuracy and speed. © 2012 Elsevier Ltd. All rights reserved.

Fuzzy community structure detection by particle competition and cooperation

Identification and classification of overlapping nodes in networks are important topics in data mining. In this paper, a network-based (graph-based) semi-supervised learning method is proposed. It is based on competition and cooperation among walking particles in a network to uncover overlapping nodes by generating continuous-valued outputs (soft labels), corresponding to the levels of membership from the nodes to each of the communities. Moreover, the proposed method can be applied to detect overlapping data items in a data set of general form, such as a vector-based data set, once it is transformed to a network. Usually, label propagation involves risks of error amplification. In order to avoid this problem, the proposed method offers a mechanism to identify outliers among the labeled data items, and consequently prevents error propagation from such outliers. Computer simulations carried out for synthetic and real-world data sets provide a numeric quantification of the performance of the method. © 2012 Springer-Verlag.

Multilayer perceptron neural networks training through charged system search and its Application for non-technical losses detection

The non-technical loss is not a problem with trivial solution or regional character and its minimization represents the guarantee of investments in product quality and maintenance of power systems, introduced by a competitive environment after the period of privatization in the national scene. In this paper, we show how to improve the training phase of a neural network-based classifier using a recently proposed meta-heuristic technique called Charged System Search, which is based on the interactions between electrically charged particles. The experiments were carried out in the context of non-technical loss in power distribution systems in a dataset obtained from a Brazilian electrical power company, and have demonstrated the robustness of the proposed technique against with several others nature-inspired optimization techniques for training neural networks. Thus, it is possible to improve some applications on Smart Grids. © 2013 IEEE.

Integração de Redes GNSS locais ao SIRGAS

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Previsão de carga multinodal utilizando redes neurais de regressão generalizada

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Teoria da ressonância adaptativa através da linguagem Java para detecção e classificação de e-mails indesejados

Pós-graduação em Engenharia Elétrica - FEIS

Análise de eventos de segurança em redes de computadores utilizando detecção de novidade

Pós-graduação em Ciência da Computação - IBILCE

Políticas públicas educacionais: prática musical na escola pública paulista

Pós-graduação em Música - IA