Pattern classification with mixtures of weighted least-squares support vector machine experts

Support Vector Machine (SVM) classifiers are high-performance classification models devised to comply with the structural risk minimization principle and to properly exploit the kernel artifice of nonlinearly mapping input data into high-dimensional feature spaces toward the automatic construction of better discriminating linear decision boundaries. Among several SVM variants, Least-Squares SVMs (LS-SVMs) have gained increased attention recently due mainly to their computationally attractive properties coming as the direct result of applying a modified formulation that makes use of a sum-squared-error cost function jointly with equality, instead of inequality, constraints. In this work, we present a flexible hybrid approach aimed at augmenting the proficiency of LS-SVM classifiers with regard to accuracy/generalization as well as to hyperparameter calibration issues. Such approach, named as Mixtures of Weighted Least-Squares Support Vector Machine Experts, centers around the fusion of the weighted variant of LS-SVMs with Mixtures of Experts models. After the formal characterization of the novel learning framework, simulation results obtained with respect to both binary and multiclass pattern classification problems are reported, ratifying the suitability of the novel hybrid approach in improving the performance issues considered.     

Brazilian articles in international journals on Limnology

Summary  We assessed the contribution of Brazilian limnologists (freshwater ecologists) in international journals in the period 1970-2004. Brazilian contribution was low and regular in the 1970's, but increased steeply after 1980 with no signs of stabilization until the present. Articles authored by Brazilians tend to be less cited than articles authored by non-Brazilians, although this difference is reduced in co-authored articles with international researchers.  Brazilian articles are not distributed homogenously among the sub-areas of Limnology, but present some biases that can be explained by intellectual legacy. Brazil has invested since the 1970's in establishing postgraduate courses in Brazil and in the last years has turned the focus to a better qualification of these courses. We believe these are the main reasons for the conspicuous development of Brazilian Limnology.     

A two-pronged approach for springback variability assessment using sparse polynomial chaos expansion and multi-level simulations

In this study, we show that stochastic analysis of metal forming process requires both a high precision and low cost numerical models in order to take into account very small perturbations on inputs (physical as well as process parameters) and to allow for numerous repeated analysis in a reasonable time. To this end, an original semi-analytical model dedicated to plain strain deep drawing based on a Bending-Under-Tension numerical model (B-U-T model) is used to accurately predict the influence of small random perturbations around a nominal solution estimated with a full scale Finite Element Model (FEM). We introduce a custom sparse variant of the Polynomial Chaos Expansion (PCE) to model the propagation of uncertainties through this model at low computational cost. Next, we apply this methodology to the deep drawing process of U-shaped metal sheet considering up to 8 random variables.     

Purification Processes and Market Potential of Bromelain in Brazil

Bromelain is the denomination given to the group of endoproteases obtained from members of Bromeliacea family. These enzymes have a wide range of proven applications and have been an object of study for worldwide researchers for decades. Over the years, several different downstream processes were studied in order to determine which technique would be worthwhile to be implemented in Brazil and provide the national market with such product. The objective of the present study is to relate the main studies in Brazil that has proven that bromelain purification can be cost-effective, in addition to the well-known benefits owned by such enzymes, and highlight the applications that create their market potential in the Brazilian market.     

Evaluation of Phase Equilibria Involving the Liquid Phase in the Hf-Si System

Metal-silicon-boron (Me-Si-B) alloys have been extensively studied due to their potential to develop high temperature structural materials. In this sense, information from accurate phase diagram becomes important. This work presents results of detailed microstructural characterization of as-cast hafnium-silicon alloys to assess the liquid composition associated to the invariant reactions of this system. The alloys were prepared by arc melting from high purity hafnium and silicon under argon atmosphere. The phases present in the alloys were identified by scanning electron microscopy and x-ray diffraction. Important discrepancies were found between the data obtained in this work and those of the most recent phase diagrams, especially the composition of the liquid phase in the following reactions: L + Hf₅Si₃ ? Hf₂Si, 32.5 at.% Si instead of 21 at.% Si, L + Hf₃Si₂ ? Hf₅Si₃, 37.0 at.% Si instead of 30.3 at.% Si and L + Hf₅Si₄ ? HfSi, 52.5 at.% Si instead of 58.3 at.% Si.     

Experimental approaches for the estimation of uncertainty in analysis of trace inorganic contaminants in foodstuffs by ICP-MS

Total Diet Studies to estimate dietary exposure to food contaminants need to evaluate laboratory measurements data variance. In this process it is critical that data from analytical methods are reliable to correctly scrutinize and compare values over time and between countries. In Europe it is widely recognized that the evaluation of measurement uncertainty is an important parameter when assessing the sources of analytical data variability. Two approaches are considered to estimate uncertainty in analytical measurement. Arsenic, Lead, Chromium and Cadmium, content in several food matrix determined by Inductively Coupled Mass Spectrometry (ICP-MS) microwave digestion assisted, are used as examples. The aim of the present research work is to compare both approaches accepted by Eurolab and GUM: Mathematical modeling to assess uncertainty components based on a classical model (bottom up) and an empirical method (top down), based on either experimental data obtained from a single laboratory validation data or inter-laboratory data from Proficiency Testing schemes. Relative expanded uncertainty calculated by both approaches agree when U (%) <20%. These values are concordant with RSD_R reported in collaborative studies of EN 15763 (2009), which were assumed as target uncertainty. The top down approach described is simple and easy to use when compared with the mathematical modeling approach providing considerable benefits to those who assess data produced by several laboratories.     

A linear programming model for cash flow management in the Brazilian construction industry

A linear programming model has been developed for optimal cash flow management addressing specific cash flow issues related to the construction industry. These include typical financial transactions, possible delays on payments, use of available credit lines, effect of changing interest rates, and budget constraints that often occur in the construction industry. A small size project from the Brazilian construction industry is provided as a case study, aiming at evaluating the potential benefits from using the model. Different changes to the basic structure of the model allow and establish the consistency of the results. Alternative formulations are suggested to deal with uncertainties, longer planning horizons, and multiple subcontractors and suppliers. In addition to the tangible financial earnings derived from the optimization process, the simple structure of the model, as a network flow and corresponding equations, provides much visual insight concerning the relationships between the external inputs and the variables of the problem. Optimal results have yet to be achieved in a real life situation, but a better view of whole cash flow management is provided when using the model.     

Disposable potentiometric citrate sensor based on polypyrrole‐doped films for indirect determination of sildenafil in pharmaceuticals formulations

A new sensitive and selective disposable potentiometric sensor based on polypyrrole (PPy) films for determination of sildenafil citrate (SC) was proposed. The pyrrole polymerization was performed in presence of citrate ions under galvanostatic conditions which resulted in a membrane of PPy doped with citrate anion at graphite pencil electrode surface. Experimental conditions (e.g., pH and conditioning time) and instrumental parameters (e.g., current density and electrical charge) were evaluated in order to reach the best potentiometric response for the proposed sensor. Under optimized conditions, the device presented a linear dynamic range (LDR) for citrate ions concentrations varying from 0.034 to 1.7 mmol L^?1 with a Nernstian slope of 57.2 mV dec^?1 and a limit of detection (LOD) of 30 µmol L^?1. The developed potentiometric sensor was applied for sildenafil citrate (SC) determination (pharmaceutical formulations) and results compared with an official spectrophotometric method indicating a good agreement for a confidence level of 95%. Effect of concomitants species on the potentiometric response of the proposed device and morphologic characterization using microscopy of atomic force (AFM) were realized. The surface roughness of PPy films (synthesized in citrate solution and chloride) showed poorly affected by changing the doping anion, probably because the polypyrrole nodules grow three‐dimensionally simultaneously. © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 43762.