Ensemble-based noise detection: noise ranking and visual performance evaluation

Noise filtering is most frequently used in data preprocessing to improve the accuracy of induced classifiers. The focus of this work is different: we aim at detecting noisy instances for improved data understanding, data cleaning and outlier identification. The paper is composed of three parts. The first part presents an ensemble-based noise ranking methodology for explicit noise and outlier identification, named Noise- Rank, which was successfully applied to a real-life medical problem as proven in domain expert evaluation. The second part is concerned with quantitative performance evaluation of noise detection algorithms on data with randomly injected noise. A methodology for visual performance evaluation of noise detection algorithms in the precision-recall space, named Viper, is presented and compared to standard evaluation practice. The third part presents the implementation of the NoiseRank and Viper methodologies in a web-based platform for composition and execution of data mining workflows. This implementation allows public accessibility of the developed approaches, repeatability and sharing of the presented experiments as well as the inclusion of web services enabling to incorporate new noise detection algorithms into the proposed noise detection and performance evaluation workflows.     

Mechanism of ZrTiO4 Synthesis by Mechanochemical Processing of TiO2 and ZrO2

High-energy ball milling initiates a solid-state reaction in an equimolar mixture of TiO₂ and ZrO₂. The first stage of ball milling induced the transformation of anatase TiO₂ to high-pressure phase TiO₂ (II), isostructural with ZrTiO₄. The formation of solid solutions monoclinic ZrO₂/TiO₂ and TiO₂ (II)/ZrO₂ was observed in the intermediate stage. Afterward, a nanosized ZrTiO₄ phase was formed in the milled product from the TiO₂ (II)/ZrO₂ solid solution. The sintering of the milled product at a temperature <1100°C was examined in situ by Raman spectroscopy. The full solid-state reaction toward ZrTiO₄ ceramic is completed at a temperature considerably lower than reported in the literature.     

Aluminium electrodeposition from AlCl3-NaCl melts on glassy carbon,platinum and gold electrodes

The electrochemical deposition and dissolution of aluminium on glassy carbon, platinum and gold electrodes in chloraluminate melts have been investigated using linear sweep voltammetry and potentiostatic pulse techniques. It was shown that deposition of aluminium on the glassy carbon electrode at low overpotentials takes place by 3-D progressive nucleation and growth, with the incorporation of atoms in the crystal lattice as the rate-determining step. At overpotentials higher than –100 mV vs Al, in the melts containing more than 52 mol % of AlCl₃, diffusion of Al₂Cl ₇ ^– , takes over the control of deposition of aluminium. Alloying of platinum and gold electrodes with aluminium from the melt occurs in the underpotential region.     

Characterisation of cut and chip behaviour for NR,SBR and BR compounds with an instrumented laboratory device

ABSTRACT

Understanding the cut and chip (CC) effect in rubber is important for successful product development for tires used in off-road or poor road conditions and for other demanding applications of rubber. This research describes a laboratory testing method for characterising the CC fracture behaviour of rubber using a device that controls and records multiple applied loads and displacements during cyclic impact to the surface of a solid rubber specimen to mimic and quantify the CC damage experienced by tire tread compounds. To study the capabilities of the instrument, three model compounds were studied that are based on carbon black reinforced compounds of common elastomers used in tire treads: natural rubber (NR), styrene-butadiene rubber (SBR), and butadiene rubber (BR). These polymers have well-established CC tendencies in field performance of tire treads, with NR exhibiting the best CC resistance followed by SBR and finally BR. The same trend was found with the rubber impact testing approach that allowed the CC behaviour to be quantified using a new physical parameter which is the CC propensity (P). The relative ranking for CC resistance for the three compounds followed the fatigue crack growth resistances of the materials but was exactly opposite to the ranking of DIN abrasion resistance. This provides evidence that CC damage from impact by mm-scale asperities and abrasion of rubber against μm-scale asperities exhibit distinct characteristics in rubber.     

Effect of temperature on the separation of some Δ9-cis-trans isomers of methyl esters of fatty acids by open tubular gas chromatography

The influence of temperature on the gas chromatographic separation ofcis-trans isomers of the methyl esters of some monounsaturated fatty acids was studied on capillary columns coated with Apiezon L, BDS and DEGS. As far as methyl oleate and methyl elaidate are concerned, the separation is better at lower temperatures on Apiezon L (180–210 C) and at higher temperatures on polyester phases (BDS, DEGS; 150–180 C). The influence of temperature on the separation ofcis-trans isomers on the three stationary phases under study is explained by the higher values of δECL/δt forcis isomers. The variation of the equivalent carbon chain length with temperature can be used for the identification ofcis-trans isomers in natural mixtures.     

Measurement of soybean fatty acids by near-infrared spectroscopy: Linear and nonlinear calibration methods

A key element of successful development of new soybean cultivars is availability of inexpensive and rapid methods for measurement of FA in seeds. Published research demonstrated applicability of NIR spectroscopy for FA profiling in oilseeds. The objectives of this study were to investigate the applicability of NIR spectroscopy for measurement of FA in whole soybeans and compare performance of calibration methods. Equations were developed using partial least squares (PLS), artificial neural networks (ANN), and support vector machines (SVM) regression methods. Validation results demonstrated that (i) equations for total saturates had the highest predictive ability (r ²=0.91–0.94) and were usable for quality assurance applications, (ii) palmitic acid models (r ²=0.80–0.84) were usable for certain research applications, and (iii) equations for stearic (r ²=0.49–0.68), oleic (r ²=0.76–0.81), linoleic (r ²=0.73–0.76), and linolenic (r ²=0.67–0.74) acids could be used for sample screening. The SVM models produced significantly more accurate predictions than those developed with PLS. ANN calibrations were not different from the other two methods. Reduction in the number of calibration samples reduced predictive ability of all equations. The rate of performance degradation of SVM models with sample reduction was the lowest.     

The participation of block copolymers in the increase in the number of tie molecules

Three-block copolymers polyamide-block-polyX-block-polyamide (where X is styrene, ethylene-co-1-butene, and others) are able to improve substantially the toughness of poly(ϵ-caprolactam). Outer blocks of these copolymers are able to take part in folding lamellae of semi-crystalline polyamide, inner blocks are not. They operate as a spacer, meaning that the end blocks of copolymers capable of folding become parts of different lamellae, or that the unfoldable part of a block copolymer is pushed out of a folding lamella with a certain probability of the remaining outer block taking part in folding another lamella. This increases the number of tie-molecules.     

The potentiometric behaviour of copolymers of methacrylic acid in water-ethanol solutions

Summary The synthesis is reported of copolymers of styrene with methacrylic acid and of methyl methacrylate with methacrylic acid by radical copolymerization, of copolymers of methyl methacrylate with methacrylic acid by partial alkaline hydrolysis of poly(methyl methacrylate), and of block copolymers of styrene with methacrylic acid. Modified titration curves of all these copolymers were recorded in water and water-ethanol solutions. In a solution containing 50 mass.% ethanol, only small differences could be observed between the potentiometric behaviour of the individual copolymers and polymethacrylic acid. Also, there were no essential differences in any of the solvents used between the potentiometric behaviour of block copolymers of styrene with methacrylic acid, on the one hand, and polymethacrylic acid, on the other. On the contrary, maxima and minima were always observed on the modified titration curves of statistical copolymers with a higher content of the hydrophobic comonomer in solutions with a high water content. Thus, using the modified titration curves, it is possible to decide whether a given copolymer is of the block or statistical type.     

Electrostatic capillary collector for in-situ spectroscopic analysis of aerosols

Identification of particulate matter is important in assessing an individual’s exposure to potentially harmful particles, such as aeroallergens, toxins, and emissions from combustion sources, which can contribute to cardio-pulmonary diseases. Efficient collection of aerosols is essential for aerosol exposure studies such as analysis of chemical and biological components. We present the design and evaluation of a capillary collector that collects PM_2.5 onto the outer surface of a capillary for in-situ spectroscopic analysis. The capillary collector uses a needle-to-ring corona generator to charge particles; the electric field between a cylinder and a wire inserted into the bore of a capillary is used to collect the charged particles. Corona and repelling voltages are optimized for maximum collection of ambient PM_2.5 particles and fluorescent polystyrene latex microspheres in the PM_2.5 size range, on the capillary. The capillary collection efficiency of ambient PM_2.5 at 3 slpm operating flow rate and optimal operating voltages is 63%. Fluorescence spectroscopy is used to quantify the collection of polystyrene latex microspheres. The fluorescence-based capillary collection efficiency is in close agreement with the capillary collection efficiency of ambient PM_2.5. The collection and analysis methodology can be used to develop a compact, low-cost sensor for in-situ spectroscopic analysis of aerosols to determine their chemical composition for source apportionment.

Copyright © 2019 American Association for Aerosol Research     

Crystallization behavior of polymer/montmorillonite nanocomposites. Part II. Intercalated poly(ε-caprolactone)/montmorillonite nanocomposites

The combination of surfactant modified montmorillonite (MMT) silicate layers, poly(ε-caprolactone) (PCL) and the adopted melt processing procedure results in intercalated nanocomposites in which the silicate layers act as nucleating agents for the crystallization of the PCL matrix and by which in turn the overall crystallization rate increases. At a sufficiently high MMT concentration and degree of supercooling the polymer-swollen silicate layer stacks disturb crystal growth, resulting in a decrease in the overall crystallization rate. Simultaneous, time resolved, synchrotron small and wide angle X-ray scattering experiments reveal that—when the retarding effect is absent at a sufficiently high temperature—the final semicrystalline structures of pure PCL and its nanocomposites are identical. The poorer nucleation in the case of pure PCL, however, results in a time wise smearing of primary and secondary crystallization whereas in the nanocomposites these events are well separated due to a nucleation induced, efficient and rapid primary crystallization. Secondary crystallization involves the insertion of new lamellar crystals in between the already existing ones.