Robust multiclass kernel-based classifiers

In this research, a robust optimization approach applied to multiclass support vector machines (SVMs) is investigated. Two new kernel based-methods are developed to address data with input uncertainty where each data point is inside a sphere of uncertainty. The models are called robust SVM and robust feasibility approach model (Robust-FA) respectively. The two models are compared in terms of robustness and generalization error. The models are compared to robust Minimax Probability Machine (MPM) in terms of generalization behavior for several data sets. It is shown that the Robust-SVM performs better than robust MPM.     

Synthesis and characterization of magnesium doped lead titanate

Magnesium (Mg) doped lead titanate powders (PT: Mg) ceramics, with various Mg contents (0, 2.5, 5, 7.5, 10, 12.5, 15, 17.5, 20, and 30 mole %), are prepared by the sol‐gel method and characterized by X‐Ray diffraction (XRD), scanning electron microscopy (SEM) and Raman spectroscopy. XRD results show that the magnesium diminishes the lattice tetragonality of PT:Mg. The average particle size and morphology were studied by SEM. The investigations done by Raman spectroscopy allow the analyzis of the substitution mechanism behavior related to the Mg incorporation into PT structure. A concentration threshold appears around 10% of Mg corresponding in a change of the compensation process with Mg²⁺ ions replacing Pb²⁺ ions below and both Pb²⁺ and Ti⁴⁺ above. (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

The inversion of acoustical impedance profile by methods of characteristics

The inverse problem for the one-dimensional propagation of waves through a medium with an unknown impedance profile is considered. From the reflected pulses caused by an excitation of short duration, this profile is reconstructed by applying the method of characteristics with finite difference approximations. This leads to a simple and fast algorithm, which is demonstrated by the numerical examples in the paper.     

Inductively coupled plasma mass spectrometry for the sequential determination of trace metals in sea water after electrothermal vaporization of their dithiocarbamate complexes in methyl isobutyl ketone

 A sensitive method has been developed for the sequential determination of V, Mn, Fe, Co, Ni, Cu, Zn, Mo and Sb in sea water using inductively coupled plasma mass spectrometry (ICP-MS) after electrothermal vaporization of their dithiocarbamate complexes in methyl isobutyl ketone (MIBK). After complexion with sodium diethyldithiocarbamate (NaDDTC), all trace analyte elements were simultaneously separated from sea water matrix and concentrated 20 fold in a single extract of MIBK, followed by introduction of 10 μL of the extract into argon plasma using a pyrolytic graphite rod electrothermal vaporizer (ETV). Sensitivity enhancement due to chemical modification using a mixed modifier of Pd(NO₃)₂-Mg(NO₃)₂ was observed for all the elements. The limits of detection ranged from 2 ng/L for Co to 329 ng/L for V. For replicate determinations of the elements in sea water, the repeatability was within 10% (as a coefficient variation), except for V (12.8%). The recovery test performed on a sea water sample resulted in a range value from 87% for Sb to 119% for V. The method has been successfully applied to sea water samples collected from the surface to the depth of 5000 m at a sampling station in the northwest Pacific Ocean. Received: 1 July 1996/Revised: 24 September 1996/Accepted: 25 September 1996     

Learning networks in rainfall estimation

This paper utilizes Artificial Neural Networks (ANNs), standard Support Vector Regression (SVR), Least-Squares Support Vector Regression (LS-SVR), linear regression (LR) and a rain rate (RR) formula that meteorologists use, to estimate rainfall. A unique source of ground truth rainfall data is the Oklahoma Mesonet. With the advent of the WSR-88D network of radars data mining is feasible for this study. The reflectivity measurements from the radar are used as inputs for the techniques tested. LS-SVR generalizes better than ANNs, linear regression and a rain rate formula in rainfall estimation and for rainfall detection, SVR has a better performance than the other techniques.Mathematics Subject Classification: 30C40     

A versatile photoacoustic spectrometer for sensitive trace-gas analysis in the mid-infrared wavelength region (5.1–8.0 and 2.8–4.1 μm)

A versatile CO laser-based photoacoustic spectrometer is presented equipped with three photoacoustic cells placed inside the laser cavity. The newly designed CO laser can operate both in the Δv=1 and the Δv=2 modes (5.1–8.0 μm and 2.8–4.1 μm) on 400 laser lines. Improved laser operation originating from a better cooling of the gas discharge was evidenced by a shift of the laser output power to lower J-values. Due to the wide emission range of the source, many molecules of biological and atmospheric interest, including methane and ethane, can be detected with sensitivities typically at the (sub)ppb level. Measurement of the respiration of a cockroach showed that the spectrometer is not only sensitive, but also has a good time response (8 s at a flow rate of 10 l/h). Received: 3 April 2002 / Revised version: 14 June 2002 / Published online: 21 August 2002 RID="*" ID="*"Current address: Soegijapranata Catholic University, Department of Food Technology, Jalan Pawiyatan Luhur IV/1 Bendan Duwur, P.O. Box 8033/SM, Semarang 50234, Indonesia RID="**" ID="**"Current address: Sanata Dharma University, FMIPA, Kampus III, Paingan, Maguwoharjo, Depok, Sleman, Tromol Pos 29, Yogyakarta 55002, Indonesia RID="***" ID="***"Corresponding author. Fax: +31-24/365-3311, E-mail: fransh@ sci.kun.nl     

Mechanical behaviour of MWCNTs reinforced electrospun nanofibres

Abstract

The nanoscale dimension of electrospun polymeric nanofibres produced by electrospinning are highly captivating, yet facing limitation of resisting external forces due to the weak tensile properties. Carbon nanotubes providing tremendous toughness due to extraordinary strong sp² bonding network of carbon atoms in honeycomb lattice structure, augmented the physical resistant strength and is easily recover to its original state after load is removed. This study reports the performance of multi-walled carbon nanotubes (MWCNTs) as filler in the electrospinning of poly (_L-lactide)-co-ε-caprolactone) (PLCL) composite nanofibres. Voltage of 10?kV is applied to the spinning solution mixture of 11?wt% (w/v) PLCL and MWCNTs, yielded nanofibres having diameters less than 400?nm. Results obtained showed the formation of composite nanofibres with tailored tensile behavior by modifying the content of MWCNTs. The addition of MWCNTs improved the tensile properties of resultant composite nanofibres, signified by tensile strength of 5.82 to 15.95?MPa, which were obtained using 0.1 to 1.0?wt% of MWCNTs. The structural integrity of nanofibres mats were retained in phosphate buffer saline (PBS) medium. Scanning Electron Microscopy (SEM) micrographs revealed the minimal of fiber deformation over 30?days of incubation and are closely identical to the initial diameter of as-spun fiber.     

Comparative analysis of long-haul system based on SSB modulation utilising dual parallel Mach–Zehnder modulators

In this paper, we have proposed a long-haul optical transmission system, based on a single sideband (SSB) modulation scheme. Analytical and simulation models have been developed, optimised and demonstrated for the proposed SSB system configurations. The SSB modulation scheme was proposed to overcome dispersion in the fibre. We have shown that the related link losses can be minimized by increasing the quality of the optical signal at the modulation. We have optimised the radio over fibre configuration scheme based on dual parallel dual drive Mach–Zehnder Modulator, thereby increasing transmission length of the fibre. With the proposed SSB, by suppressing some of the harmonics and cancelling one of the sidebands, we have halved the RF power fading and interference. The developed analytical (theoretical/mathematical) model agrees very well with the simulation results using two (both) different commercial simulation tools. The optical signal is boosted while minimizing the number of repeaters. We report a SSB configuration, compensation and amplification with individual spans of 150 km, by extending the length of the link up to 3250 km. The proposed system configuration exhibits high performance with less complexity and lower cost.     

Genus Smenospongia: Untapped Treasure of Biometabolites—Biosynthesis,Synthesis, and Bioactivities

Marine sponges continue to attract remarkable attention as one of the richest pools of bioactive metabolites in the marine environment. The genus Smenospongia (order Dictyoceratida, family Thorectidae) sponges can produce diverse classes of metabolites with unique and unusual chemical skeletons, including terpenoids (sesqui-, di-, and sesterterpenoids), indole alkaloids, aplysinopsins, bisspiroimidazolidinones, chromenes, γ-pyrones, phenyl alkenes, naphthoquinones, and polyketides that possessed diversified bioactivities. This review provided an overview of the reported metabolites from Smenospongia sponges, including their biosynthesis, synthesis, and bioactivities in the period from 1980 to June 2022. The structural characteristics and diverse bioactivities of these metabolites could attract a great deal of attention from natural-product chemists and pharmaceuticals seeking to develop these metabolites into medicine for the treatment and prevention of certain health concerns.