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.     

A new set distance and its application to shape registration

We propose a new distance measure, called Complement weighted sum of minimal distances, between finite sets in ${\mathbb Z }^n$ and evaluate its usefulness for shape registration and matching. In this set distance the contribution of each point of each set is weighted according to its distance to the complement of the set. In this way, outliers and noise contribute less to the new similarity measure. We evaluate the performance of the new set distance for registration of shapes in binary images and compare it to a number of often used set distances found in the literature. The most extensive evaluation uses a set of synthetic 2D images. We also show three examples of real problems: registering a set of 2D images extracted from synchrotron radiation micro-computed tomography (SR $\upmu $ CT) volumes depicting bone implants; the difficult multi-modal registration task of finding the exact location of a 2D slice of a bone implant, as imaged by a light microscope, within a 3D SR $\upmu $ CT volume of the same implant; and finally recognition of handwritten characters. The evaluation shows that our new set distance performs well for all tasks and outperforms the other observed distance measures in most cases. It is therefore useful in many image registration and shape comparison tasks.     

HamleDT: Harmonized multi-language dependency treebank

We present HamleDT—a HArmonized Multi-LanguagE Dependency Treebank. HamleDT is a compilation of existing dependency treebanks (or dependency conversions of other treebanks), transformed so that they all conform to the same annotation style. In the present article, we provide a thorough investigation and discussion of a number of phenomena that are comparable across languages, though their annotation in treebanks often differs. We claim that transformation procedures can be designed to automatically identify most such phenomena and convert them to a unified annotation style. This unification is beneficial both to comparative corpus linguistics and to machine learning of syntactic parsing.     

Influence of anionic form on thermooxidation of TMAHP–cellulose

With the help of DTA, DTG elementary analysis of carbonized residues and ESR spectroscopy the influence of anionic form on thermooxidation of trimethylammoniumhydroxyprophyl (TMAHP)–cellulose was studied. At 300°C the percentage of carbon in carbonized residue thermolyzed in oxidative atmosphere is higher than for the sample degraded in inert atmosphere. The percentage of hydrogen decreases simultaneously. The concentration of free radicals in thermolyzed residue also increases due to the presence of oxygen. We propose that oxygen is abstracting the hydrogen atoms from polysaccharide and unpaired electrons on carbon atoms are produced. At 400°C the percentage of carbon in residues prepared at inert atmosphere is higher than for residue formed at oxidative atmosphere. Also the concentration of free radicals in thermolyzed residues obtained in inert atmosphere is greater than for those from oxidative ones. That is why suppose that at this temperature oxygen is bonded to polysaccharide residue and free radicals are terminated. From the semiquantitative DTA we can make the following sequence of samples according to their increasing thermooxidative effect: unmodified cellulose < A–HSO < A–Br^? < A–I^? < A–NO < A–H₂PO < A–CH₃COO^? < A–HCO < A–F^? < A–Cl^?1 < A–OH^p?.     

Theory of porous electrodes—XI. The positive plate of the lead-acid battery

Partial differential equations describing the transport of mass and electricity in the pores of the positive electrode of a lead acid battery were derived. The theory is based on exact transport equations and on the assumption that the solid porous matrix has a metallic conductivity. Volume changes in both phases are taken into account. Numerical solutions obtained on a computer are presented for the case where the influence of electrolyte between the electrodes can be neglected. The solutions depend on the product of electrode thickness times current density, and on the initial porosity.     

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.     

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.     

Microstructure and physical properties of transparent thoria-yttria ceramics

The investigation of microstructure, some electrical and optical parameters of transparent ThO₂-15 mol % yo_1.5 ceramics has shown: (a) a very low porosity, close to zero; (b) a higher electrical conductivity than is usual for ThO₂ ceramics of the same composition, with the activation energy 1.12 eV; (c) a relative permittivity _r 33 under the given conditions; (d) confirmation of a single-phase fluorite-type cubic structure by means of Raman and infrared spectra.     

Fusion of smartphone sensor data for classification of daily user activities

Multimedia Tools and Applications - New mobile applications need to estimate user activities by using sensor data provided by smart wearable devices and deliver context-aware solutions to users...