Learning kernels for support vector machines with polynomial powers of sigmoid

In the pattern recognition research field, Support Vector Machines (SVM) have been an effectiveness tool for classification purposes, being successively employed in many applications. The SVM input data is transformed into a high dimensional space using some kernel functions where linear separation is more likely. However, there are some computational drawbacks associated to SVM. One of them is the computational burden required to find out the more adequate parameters for the kernel mapping considering each non-linearly separable input data space, which reflects the performance of SVM. This paper introduces the Polynomial Powers of Sigmoid for SVM kernel mapping, and it shows their advantages over well-known kernel functions using real and synthetic datasets.

Epsilon(b) constrainsts on self-couplings of vector bosons

We analyze the constraints on possible anomalous contributions to the W+W-Z vertex coming from non-universal radiative corrections to the Z → bb̄ width. We parametrize these corrections in terms of ∈b and use the LEP data to establish the allowed values for the anomalous triple couplings. We examine all CP conserving effective operators that exhibit SU(2)L × U(1)Y gauge invariance and do not give any tree level contribution to the present experimental observables. For some of these operators our constraints are comparable with the bounds coming from a global fit of the oblique parameters, which evidences the increasing relevance of the precise measurement of the b-quark parameters at LEP for the search of new physics.

Evidence for temporal population replacement and the signature of ecological adaptation in a major Neotropical malaria vector in Amazonian Peru

The major Neotropical malaria vector, Anopheles darlingi, was reintroduced into the Iquitos, Loreto, Peru area during the early 1990s, where it displaced other anophelines and caused a major malaria epidemic. Since then, case numbers in Loreto have fluctuated, but annual increases have been reported since 2012. The population genetic structure of An. darlingi sampled before and after the introduction of long-lasting insecticidal nets (LLINs) was investigated to test the hypothesis of temporal population change (2006 vs. 2012). Current samples of An. darlingi were used to test the hypothesis of ecological adaptation to human modified (highway) compared with wild (riverine) habitat, linked to forest cover. In total, 693 An. darlingi from nine localities in Loreto, Peru area were genotyped using 13 microsatellite loci. To test the hypothesis of habitat differentiation in An. darlingi biting time patterns, HBR and EIR, four collections of An. darlingi from five localities (two riverine and three highway) were analysed. Analyses of microsatellite loci from seven (2006) and nine settlements (2012-2014) in the Iquitos area detected two distinctive populations with little overlap, although it is unclear whether this population replacement event is associated with LLIN distribution or climate. Within the 2012-2014 population two admixed subpopulations, A and B, were differentiated by habitat, with B significantly overrepresented in highway, and both in near-equal proportions in riverine. Both subpopulations had a signature of expansion and there was moderate genetic differentiation between them. Habitat and forest cover level had significant effects on HBR, such that Plasmodium transmission risk, as measured by EIR, in peridomestic riverine settlements was threefold higher than in peridomestic highway settlements. HBR was directly associated with available host biomass rather than forest cover. A population replacement event occurred between 2006 and 2012-2014, concurrently with LLIN distribution and a moderate El Niño event, and prior to an increase in malaria incidence. The likely drivers of this replacement cannot be determined with current data. The present-day An. darlingi population is composed of two highly admixed subpopulations, which appear to be in an early stage of differentiation, triggered by anthropogenic alterations to local habitat.

A hybrid method for the solution of a sparse power system matrices on vector computers

This paper describes a methodology for solving a linear system of equations on vector computer. The methodology combines direct and inverse factors. The decomposition and implementation of the direct solution in a CRAY Y-MPZE/232, and the performance results are discussed.

Effects of a mixed vector-scalar kink-like potential for spinless particles in two-dimensional space time

The intrinsically relativistic problem of spinless particles subject to a general mixing of vector and scalar kink- like potentials (similar to tanh gamma x) is investigated. The problem is mapped into the exactly solvable Sturm - Liouville problem with the Rosen - Morse potential and exact bounded solutions for particles and antiparticles are found. The behavior of the spectrum is discussed in some detail. An apparent paradox concerning the uncertainty principle is solved by recurring to the concept of effective Compton wavelength.

A simulation tool for transient stability analysis on vector computers

This paper deals with transient stability analysis based on time domain simulation on vector processing. This approach requires the solution of a set of differential equations in conjunction of another set of algebraic equations. The solution of the algebraic equations has presented a scalar as sequential set of tasks, and the solution of these equations, on vector computers, has required much more investigations to speedup the simulations. Therefore, the main objective of this paper has been to present methods to solve the algebraic equations using vector processing. The results, using a GRAY computer, have shown that on-line transient stability assessment is feasible.

Managing Large Data Sets Using Support Vector Machines

Hundreds of Terabytes of CMS (Compact Muon Solenoid) data are being accumulated for storage day by day at the University of Nebraska-Lincoln, which is one of the eight US CMS Tier-2 sites. Managing this data includes retaining useful CMS data sets and clearing storage space for newly arriving data by deleting less useful data sets. This is an important task that is currently being done manually and it requires a large amount of time. The overall objective of this study was to develop a methodology to help identify the data sets to be deleted when there is a requirement for storage space. CMS data is stored using HDFS (Hadoop Distributed File System). HDFS logs give information regarding file access operations. Hadoop MapReduce was used to feed information in these logs to Support Vector Machines (SVMs), a machine learning algorithm applicable to classification and regression which is used in this Thesis to develop a classifier. Time elapsed in data set classification by this method is dependent on the size of the input HDFS log file since the algorithmic complexities of Hadoop MapReduce algorithms here are O(n). The SVM methodology produces a list of data sets for deletion along with their respective sizes. This methodology was also compared with a heuristic called Retention Cost which was calculated using size of the data set and the time since its last access to help decide how useful a data set is. Accuracies of both were compared by calculating the percentage of data sets predicted for deletion which were accessed at a later instance of time. Our methodology using SVMs proved to be more accurate than using the Retention Cost heuristic. This methodology could be used to solve similar problems involving other large data sets.

Fuzzy linguistic model for bioelectrical impedance vector analysis

Backgrounds Ea aims: The boundaries between the categories of body composition provided by vectorial analysis of bioimpedance are not well defined. In this paper, fuzzy sets theory was used for modeling such uncertainty. Methods: An Italian database with 179 cases 18-70 years was divided randomly into developing (n = 20) and testing samples (n = 159). From the 159 registries of the testing sample, 99 contributed with unequivocal diagnosis. Resistance/height and reactance/height were the input variables in the model. Output variables were the seven categories of body composition of vectorial analysis. For each case the linguistic model estimated the membership degree of each impedance category. To compare such results to the previously established diagnoses Kappa statistics was used. This demanded singling out one among the output set of seven categories of membership degrees. This procedure (defuzzification rule) established that the category with the highest membership degree should be the most likely category for the case. Results: The fuzzy model showed a good fit to the development sample. Excellent agreement was achieved between the defuzzified impedance diagnoses and the clinical diagnoses in the testing sample (Kappa = 0.85, p < 0.001). Conclusions: fuzzy linguistic model was found in good agreement with clinical diagnoses. If the whole model output is considered, information on to which extent each BIVA category is present does better advise clinical practice with an enlarged nosological framework and diverse therapeutic strategies. (C) 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Geochemical signatures of metasedimentary rocks of high-pressure granulite facies and their relation with partial melting: Carvalhos Klippe, Southern Brasilia Belt, Brazil

High-grade metasedimentary rocks can preserve geochemical signatures of their sedimentary protolith if significant melt extraction did not occur. Retrograde reaction textures provide the main evidence for trapped melt in the rock fabrics. Carvalhos Klippe rocks in Southern Brasilia Orogen, Brazil, present a typical high-pressure granulite assemblage with evidence of mica breakdown partial melting (Ky + Grt + Kfs +/- Bt +/- Rt). The metamorphic peak temperatures obtained by Zr-in-Rt and ternary feldspar geothermometers are between 850 degrees C and 900 degrees C. The GASP bane peak pressure obtained using grossular rich garnet core is 16 kbar. Retrograde reaction textures in which the garnet crystals are partially to totally replaced by Bt + Qtz +/- Fsp intergrowths are very common in the Carvalhos Klippe rocks. These reactions are interpreted as a result of interactions between residual phases and trapped melt during the retrograde path. In the present study the geochemical signatures of three groups of Carvalhos Klippe metasedimentary rocks are analysed. Despite the high metamorphic grade these three groups show well-defined geochemical features and their REE patterns are similar to average compositions of post-Archean sedimentary rocks (PAAS, NASC). The high-pressure granulite facies Grt-Bt-Pl gneisses with immature arenite (wacke, arkose or lithic-arenite) geochemical signatures present in the Carvalhos Klippe are compared to similar rocks in amphibolite facies from the same tectonic framework (Andrelandia Nappe System). The similar geochemical signatures between Grt-Bt-Pl gneisses metamorphosed in high-pressure granulite facies and Grt-Bt-Pl-Qtz schists from the Andrelandia and Liberdade Nappes, with minimal to absent melting conditions, are suggestive of low rates of melt extraction in these high-grade rocks. The rocks with pelitic compositions most likely had higher melt extraction and even under such circumstances nevertheless tend to show REE patterns similar to average compositions of post-Archean sedimentary rocks (PAAS, NASC). (C) 2012 Elsevier Ltd. All rights reserved.

Solvability in the large for a class of complex vector fields on the cylinder

This work deals with global solvability of a class of complex vector fields of the form L = partial derivative/partial derivative t + (a(x, t)+ ib(x, t))partial derivative/partial derivative x, where a and b are real-valued C-infinity functions, defined on the cylinder Omega = R x S-1. Relatively compact (Sussmann) orbits are allowed. The connection with Malgrange's notion of L-convexity for supports is investigated. (C) 2011 Elsevier Masson SAS. All rights reserved.

Solvability near the characteristic set for a special class of complex vector fields

This work deals with the solvability near the characteristic set Sigma = {0} x S-1 of operators of the form L = partial derivative/partial derivative t+(x(n) a(x)+ ix(m) b(x))partial derivative/partial derivative x, b not equivalent to 0 and a(0) not equal 0, defined on Omega(epsilon) = (-epsilon, epsilon) x S-1, epsilon > 0, where a and b are real-valued smooth functions in (-epsilon, epsilon) and m >= 2n. It is shown that given f belonging to a subspace of finite codimension of C-infinity (Omega(epsilon)) there is a solution u is an element of L-infinity of the equation Lu = f in a neighborhood of Sigma; moreover, the L-infinity regularity is sharp.

Amazonian malaria: Asymptomatic human reservoirs, diagnostic challenges, environmentally driven changes in mosquito vector populations, and the mandate for sustainable control strategies

Across the Americas and the Caribbean, nearly 561,000 slide-confirmed malaria infections were reported officially in 2008. The nine Amazonian countries accounted for 89% of these infections; Brazil and Peru alone contributed 56% and 7% of them, respectively. Local populations of the relatively neglected parasite Plasmodium vivax, which currently accounts for 77% of the regional malaria burden, are extremely diverse genetically and geographically structured. At a time when malaria elimination is placed on the public health agenda of several endemic countries, it remains unclear why malaria proved so difficult to control in areas of relatively low levels of transmission such as the Amazon Basin. We hypothesize that asymptomatic parasite carriage and massive environmental changes that affect vector abundance and behavior are major contributors to malaria transmission in epidemiologically diverse areas across the Amazon Basin. Here we review available data supporting this hypothesis and discuss their implications for current and future malaria intervention policies in the region. Given that locally generated scientific evidence is urgently required to support malaria control interventions in Amazonia, we briefly describe the aims of our current field-oriented malaria research in rural villages and gold-mining enclaves in Peru and a recently opened agricultural settlement in Brazil. (C) 2011 Elsevier B.V. All rights reserved.

Spectral Induced Polarization (SIP) signatures of clayey soils containing toluene

We performed laboratory experiments to investigate the sensitivity of the Spectral Induced Polarization (SIP) method to toluene contamination in clayey soils. We used mixtures of quartzitic sand and montmorillonite as soil samples, artificially contaminated with varying amounts of toluene. Care was taken to quantify the experimental uncertainty resulting from packing since such effects must be quantified if variations in SIP signatures between samples are to be reliably interpreted in terms of the effects of hydrocarbon concentration. The SIP response of all samples following addition of toluene was monitored for a period of 40 days following sample preparation. Stepwise regression was used to examine the statistical significance of correlations between (i) clay content and (ii) toluene concentration and SIP parameters. Both single-frequency real and imaginary conductivity measurements, along with the integral chargeability, normalized chargeability, DC conductivity and time constant obtained from a Debye decomposition fitting, were examined in this regression analysis. The SIP measurements show a clear time dependence following sample preparation, indicating that samples containing toluene may take significant time to reach an equilibrium electrical response. SIP measurements are significantly related to toluene content shortly after sample preparation, when the expected dependence of SIP on clay concentration is apparently suppressed. However, for the state of electrical equilibrium after 40 days (interpreted to indicate surface chemistry at equilibrium) there is no significant relation between SIP measurements and toluene content; instead SIP measurements are then significantly correlated with clay concentration. The total chargeability, normalized chargeability and relaxation time obtained from the Debye decomposition show no correlation with toluene content, indicating that this procedure, which likely integrates over multiple mechanisms, may not be suitable for understanding relationships between SIP and hydrocarbon contamination. We find only small low-frequency polarization signals observed in relation to toluene concentration (2 mrad at 0.01 Hz), which initially decreases the interfacial polarization. Unlike earlier works, our results do not support the use of the SIP method as a tool for monitoring toluene contamination in clay soils.

Globally solvable systems of complex vector fields

We consider a class of involutive systems of n smooth vector fields on the n + 1 dimensional torus. We obtain a complete characterization for the global solvability of this class in terms of Liouville forms and of the connectedness of all sublevel and superlevel sets of the primitive of a certain 1-form in the minimal covering space.

Bioelectrical impedance vector analysis (BIVA) in stable preterm newborns

Objective: To observe the behavior of the plotted vectors on the RXc (R - resistance - and Xc - reactance corrected for body height/length) graph through bioelectrical impedance analysis (BIVA) and phase angle (PA) values in stable premature infants, considering the hypothesis that preterm infants present vector behavior on BIVA suggestive of less total body water and soft tissues, compared to reference data for term infants. Methods: Cross-sectional study, including preterm neonates of both genders, in-patients admitted to an intermediate care unit at a tertiary care hospital. Data on delivery, diet and bioelectrical impedance (800 mA, 50 kHz) were collected. The graphs and vector analysis were performed with the BIVA software. Results: A total of 108 preterm infants were studied, separated according to age (< 7 days and >= 7 days). Most of the premature babies were without the normal range (above the 95% tolerance intervals) existing in literature for term newborn infants and there was a tendency to dispersion of the points in the upper right quadrant, RXc plan. The PA was 4.92 degrees (+/- 2.18) for newborns < 7 days and 4.34 degrees (+/- 2.37) for newborns >= 7 days. Conclusion: Premature infants behave similarly in terms of BIVA and most of them have less absolute body water, presenting less fat free mass and fat mass in absolute values, compared to term newborn infants.