Efficient Development of Lexical Language Resources and their Representation

Statistical approaches in speech technology, whether used for statistical language models, trees, hidden Markov models or neural networks, represent the driving forces for the creation of language resources (LR), e.g., text corpora, pronunciation and morphology lexicons, and speech databases. This paper presents a system architecture for the rapid construction of morphologic and phonetic lexicons, two of the most important written language resources for the development of ASR (automatic speech recognition) and TTS (text-to-speech) systems. The presented architecture is modular and is particularly suitable for the development of written language resources for inflectional languages. In this paper an implementation is presented for the Slovenian language. The integrated graphic user interface focuses on the morphological and phonetic aspects of language and allows experts to produce good performances during analysis. In multilingual TTS systems, many extensive external written language resources are used, especially in the text processing part. It is very important, therefore, that representation of these resources is time and space efficient. It is also very important that language resources for new languages can be easily incorporated into the system, without modifying the common algorithms developed for multiple languages. In this regard the use of large external language resources (e.g., morphology and phonetic lexicons) represent an important problem because of the required space and slow look-up time. This paper presents a method and its results for compiling large lexicons, using examples for compiling German phonetic and morphology lexicons (CISLEX), and Slovenian phonetic (SIflex) and morphology (SImlex) lexicons, into corresponding finite-state transducers (FSTs). The German lexicons consisted of about 300,000 words, SIflex consisted of about 60,000 and SImlex of about 600,000 words (where 40,000 words were used for representation using finite-state transducers). Representation of large lexicons using finite-state transducers is mainly motivated by considerations of space and time efficiency. A great reduction in size and optimal access time was achieved for all lexicons. The starting size for the German phonetic lexicon was 12.53 MB and 18.49 MB for the morphology lexicon. The starting size for the Slovenian phonetic lexicon was 1.8 MB and 1.4 MB for the morphology lexicon. The final size of the corresponding FSTs was 2.78 MB for the German phonetic lexicon, 6.33 MB for the German morphology lexicon, 253 KB for SIflex and 662 KB for the SImlex lexicon. The achieved look-up time is optimal, since it only depends on the length of the input word and not on the size of the lexicon. Integration of lexicons for new languages into the multilingual TTS system is easy when using such representations and does not require any changes in the algorithms used for such lexicons.     

Soil-moisture estimation from TerraSAR-X data using neural networks

Early prediction of natural disasters like floods and landslides is essential for reasons of public safety. This can be attained by processing Synthetic-Aperture Radar (SAR) images and retrieving soil-moisture parameters. In this article, TerraSAR-X product images are investigated in combination with a water-cloud model based on the Shi semi-empirical model to determine the accuracy of soil-moisture parameter retrieval. SAR images were captured between January 2008 and September 2010 in the vicinity of the city Maribor, Slovenia, at different incidence angles. The water-cloud model provides acceptable estimated soil-moisture parameters at bare or scarcely vegetated soil areas. However, this model is too sensitive to speckle noise; therefore, a pre-processing step for speckle-noise reduction is carried out. Afterwards, self-organizing neural networks (SOM) are used to segment the areas at which the performance of this model is poor, and at the same time neural networks are also used for a more accurate approximation of model parameters’ values. Ground-truth is measured using the Pico64 sensor located on the field, simultaneously with capturing SAR images, in order to enable the comparison and validation of the obtained results. Experimental results show that the proposed method outperforms the water-cloud model accuracy over all incidence angles.     

Banks and information technology: marketability vs. relationships

This paper evaluates the impact of information technology (IT) on the operations of banks and the structure of the banking industry, including implications for stability. On the one hand, banks can focus on relationship banking and use IT developments to tailor services to individual needs and build enhanced, albeit modified, relationships with customers. On the other hand, IT better allows banks to exploit economies of scale and scope, which are most evident in transaction banking. Another manifestation of IT is via financial innovations that have enhanced marketability. Stability enters the picture because increased marketability facilitates opportunistic behavior. Together with enhanced herding behavior and changes in industry structure, this could undermine stability and augment systemic risk, calling for a regulatory overhaul.     

Practical considerations for 3-D image reconstruction using spherically symmetric volume elements

Spherically symmetric volume elements with smooth tapering of the values near their boundaries are alternatives to the more conventional voxels for the construction of volume images in the computer. Their use, instead of voxels, introduces additional parameters which enable the user to control the shape of the volume element (blob) and consequently to control the characteristics of the images produced by iterative methods for reconstruction from projection data. For images composed of blobs, efficient algorithms have been designed for the projection and discrete back-projection operations, which are the crucial parts of iterative reconstruction methods. The authors have investigated the relationship between the values of the blob parameters and the properties of images represented by the blobs. Experiments show that using blobs in iterative reconstruction methods leads to substantial improvement in the reconstruction performance, based on visual quality and on quantitative measures, in comparison with the voxel case. The images reconstructed using appropriately chosen blobs are characterized by less image noise for both noiseless data and noisy data, without loss of image resolution.     

On the possibility of direct Fourier reconstruction from divergent-beam projections

A great number of tomographic systems, especially those equipped with fast data acquisition techniques, scan the objects investigated by divergent (fan) X-ray beams. Fan-beam projection data require special reconstruction techniques to be implemented. Among reconstruction techniques from parallel projection data, the direct Fourier method (DFM) proved to be one of the most promising ones, especially for high-speed image reconstruction in the high-end 3-D and dynamic tomographic systems. The goal of this work is to answer the topical question: how would direct use of the DFM influence the quality of image reconstruction from the fan-beam data? The formula describing the error caused by such an approximation is derived. The conclusions deduced from the formula are confirmed by computer simulations. The boundary values of data acquisition geometry parameters have been estimated for the case of using the DFM without recalculating the fan-beam data.     

Alzheimer’s Disease-Associated SNP rs708727 in SLC41A1 May Increase Risk for Parkinson’s Disease: Report from Enlarged Slovak Study

SLC41A1 (A1) SNPs rs11240569 and rs823156 are associated with altered risk for Parkinson’s disease (PD), predominantly in Asian populations, and rs708727 has been linked to Alzheimer’s disease (AD). In this study, we have examined a potential association of the three aforementioned SNPs and of rs9438393, rs56152218, and rs61822602 (all three lying in the A1 promoter region) with PD in the Slovak population. Out of the six tested SNPs, we have identified only rs708727 as being associated with an increased risk for PD onset in Slovaks. The minor allele (A) in rs708727 is associated with PD in dominant and completely over-dominant genetic models (OR_D = 1.36 (1.05–1.77), p = 0.02, and OR_COD = 1.34 (1.04–1.72), p = 0.02). Furthermore, the genotypic triplet GG_(rs708727) + AG_(rs823156) + CC_(rs61822602) might be clinically relevant despite showing a medium (h ≥ 0.5) size difference (h = 0.522) between the PD and the control populations. RandomForest modeling has identified the power of the tested SNPs for discriminating between PD-patients and the controls to be essentially zero. The identified association of rs708727 with PD in the Slovak population leads us to hypothesize that this A1 polymorphism, which is involved in the epigenetic regulation of the expression of the AD-linked gene PM20D1, is also involved in the pathoetiology of PD (or universally in neurodegeneration) through the same or similar mechanism as in AD.     

Can round goby (Neogobius melanostomus) caught by rod and line be used for diet analysis?

Sample structure and diet of round gobies (Neogobius melanostomus) captured by angling (ANG) and electrofishing (EF) at adjacent sites in their native distribution range (Bulgarian Danube) were compared to determine whether ANG fish could be used for diet analysis. In total, 100 round gobies were captured, 52 through angling and 48 by electrofishing. EF fish were significantly smaller than ANG fish, though modal size was comparable. There was no significant difference in condition between EF and ANG. Sex ratios did not differ significantly from 1:1 in either sample. ANG fish had significantly more empty stomachs than EF fish (56% and 4%, respectively). Thirteen food items were recorded, with no significant difference in diversity between ANG and EF. Crustaceans dominated in ANG fish diet and molluscs in EF diet, though this may be an artefact of degree of stomach fullness. We suggest that angling alone is unsuitable for sampling fish for diet analysis, as it is highly selective for both fish size and degree of stomach fullness.     

Antimony powder-modified carbon paste electrodes for electrochemical stripping determination of trace heavy metals

A new type of the antimony electrode based on a carbon paste bulk-modified with antimony powder (Sb-CPE) is presented for the determination of cadmium(II) and lead(II) ions at the microgram-per-liter concentration level, when using square-wave anodic stripping voltammetric or stripping chronopotentiometric protocol. The Sb-CPE was prepared by mixing fine antimony powder, carbon powder, and silicon oil, thus combining typical features of the carbon paste material with specific electrochemical properties of antimony. Key-operational parameters, including the amount of antimony powder in the carbon paste mixture, effect of the deposition potential and deposition time, were optimized and electroanalytical performance of the Sb-CPE in nondeaerated solution of 0.01 M hydrochloric acid (pH 2) was compared with related bismuth powder-modified carbon paste electrode (Bi-CPE) and with in situ prepared antimony film carbon paste electrode (SbF-CPE). The electrode of interest exhibited well-developed signals and highly linear calibration plots for both metal ions tested. In addition, favorable limits of detection were achieved; namely: 1.4 μg L⁻¹ for Cd(II) and 0.9 μg L⁻¹ for Pb(II). The applicability of the new electrode was demonstrated on the analysis of tap water (spiked sample). Besides voltammetric measurements, the Sb-CPE was preliminary tested also under chronopotentiometric stripping mode in deoxygenated solutions, revealing also an excellent electroanalytical performance.     

Multi-label feature ranking with ensemble methods

In this paper, we propose three ensemble-based feature ranking scores for multi-label classification (MLC), which is a generalisation of multi-class classification where the classes are not mutually exclusive. Each of the scores (Symbolic, Genie3 and Random forest) can be computed from three different ensembles of predictive clustering trees: Bagging, Random forest and Extra trees. We extensively evaluate the proposed scores on 24 benchmark MLC problems, using 15 standard MLC evaluation measures. We determine the ranking quality saturation points in terms of the ensemble sizes, for each ranking-ensemble pair, and show that quality rankings can be computed really efficiently (typically 10 or 50 trees suffice). We also show that the proposed feature rankings are relevant and determine the most appropriate ensemble method for every feature ranking score. We empirically prove that the proposed feature ranking scores outperform current state-of-the-art methods in the quality of the rankings (for the majority of the evaluation measures), and in time efficiency. Finally, we determine the best performing feature ranking scores. Taking into account the quality of the rankings first and—in the case of ties—time efficiency, we identify the Genie3 feature ranking score as the optimal one.

     

Room temperature deposition of freestanding BaTiO3 films: temperature-induced irreversible structural and chemical relaxation

The room temperature aerosol deposition method is especially promising for the rapid deposition of ceramic thick films, making it interesting for functional components in energy, mobility, and telecommunications applications. Despite this, a number of challenges remain, such as an enhanced electrical conductivity and internal residual stresses in as-deposited films. In this work, a novel technique that integrates a sacrificial water-soluble buffer layer was used to fabricate freestanding ceramic thick films, which allows for direct observation of the film without influence of the substrate or prior thermal treatment. Here, the temperature-dependent chemical and structural relaxation phenomena in freestanding BaTiO₃ films were directly investigated by characterizing the thermal expansion properties and temperature-dependent crystal structure as a function of oxygen partial pressure, where a clear nonlinear, hysteretic contraction was observed during heating, which is understood to be influenced by lattice defects. As such, aliovalent doping and atmosphere-dependent annealing experiments were used to demonstrate the influence of local chemical redistribution and oxygen vacancies on the thermal expansion, leading to insight into the origin of the high room temperature conductivity of as-deposited films as well as greater insight into the influence of the induced chemical, structural, and microstructural changes in room temperature deposited functional ceramic thick films.

Graphical abstract