A prosodic phrasing model for a Korean text-to-speech synthesis system

This paper presents a prosodic phrasing model for Korean to be used in a text-to-speech synthesis (TTS) system. Read text corpora were morpho-syntactically parsed and prosodically labeled following the Penn Korean Treebank (Han, Chunghye, Ko, Eon-Suk, Yi, Heejong, Palmer, M., 2002. Penn Korean Treebank: development and evaluation. In: Proceedings of the 16th Pacific Asian Conference on Language and Computation. Korean Society for Language and Information.) and K-ToBI prosodic labeling conventions (Sun-Ah, J., 2000. K-ToBI (Korean ToBI) labelling conventions. Version 3.1. Available from: URL <http://www.linguistics.ucla.edu/people/jun/ktobi/K-tobi.html>.), respectively. Decision trees were trained with morpho-syntactic and textual distance features to predict locations of accentual and intonational phrase breaks. Our phrasing model cross-validated on a 300-sentence corpus (6936 words or 21,436 syllables, with an average of 72 syllables or 23 words per sentence) predicted non-breaks with F = 92.4% and breaks with F = 88.0% (F = 72.8% for accentual phrase breaks and F = 71.3% for intonational phrase breaks).     

A hybrid classifier combining SMOTE with PSO to estimate 5-year survivability of breast cancer patients

In this study, we propose a set of new algorithms to enhance the effectiveness of classification for 5-year survivability of breast cancer patients from a massive data set with imbalanced property. The proposed classifier algorithms are a combination of synthetic minority oversampling technique (SMOTE) and particle swarm optimization (PSO), while integrating some well known classifiers, such as logistic regression, C5 decision tree (C5) model, and 1-nearest neighbor search. To justify the effectiveness for this new set of classifiers, the g-mean and accuracy indices are used as performance indexes; moreover, the proposed classifiers are compared with previous literatures. Experimental results show that the hybrid algorithm of SMOTE + PSO + C5 is the best one for 5-year survivability of breast cancer patient classification among all algorithm combinations. We conclude that, implementing SMOTE in appropriate searching algorithms such as PSO and classifiers such as C5 can significantly improve the effectiveness of classification for massive imbalanced data sets.     

A new approach of audio emotion recognition

A new architecture of intelligent audio emotion recognition is proposed in this paper. It fully utilizes both prosodic and spectral features in its design. It has two main paths in parallel and can recognize 6 emotions. Path 1 is designed based on intensive analysis of different prosodic features. Significant prosodic features are identified to differentiate emotions. Path 2 is designed based on research analysis on spectral features. Extraction of Mel-Frequency Cepstral Coefficient (MFCC) feature is then followed by Bi-directional Principle Component Analysis (BDPCA), Linear Discriminant Analysis (LDA) and Radial Basis Function (RBF) neural classification. This path has 3 parallel BDPCA + LDA + RBF sub-paths structure and each handles two emotions. Fusion modules are also proposed for weights assignment and decision making. The performance of the proposed architecture is evaluated on eNTERFACE’05 and RML databases. Simulation results and comparison have revealed good performance of the proposed recognizer.     

Performance comparison of artificial neural network and logistic regression model for differentiating lung nodules on CT scans

PurposeTo compare the diagnostic performances of artificial neural networks (ANNs) and multivariable logistic regression (LR) analyses for differentiating between malignant and benign lung nodules on computed tomography (CT) scans.MethodsThis study evaluated 135 malignant nodules and 65 benign nodules. For each nodule, morphologic features (size, margins, contour, internal characteristics) on CT images and the patient’s age, sex and history of bloody sputum were recorded. Based on 200 bootstrap samples generated from the initial dataset, 200 pairs of ANN and LR models were built and tested. The area under the receiver operating characteristic (ROC) curve, Hosmer–Lemeshow statistic and overall accuracy rate were used for the performance comparison.ResultsANNs had a higher discriminative performance than LR models (area under the ROC curve: 0.955 ± 0.015 (mean ± standard error) and 0.929 ± 0.017, respectively, p < 0.05). The overall accuracy rate for ANNs (90.0 ± 2.0%) was greater than that for LR models (86.9 ± 1.6%, p < 0.05). The Hosmer–Lemeshow statistic for the ANNs was 8.76 ± 6.59 vs. 6.62 ± 4.03 (p > 0.05) for the LR models.ConclusionsWhen used to differentiate between malignant and benign lung nodules on CT scans based on both objective and subjective features, ANNs outperformed LR models in both discrimination and clinical usefulness, but did not outperform for the calibration.     

Genetic programming for epileptic pattern recognition in electroencephalographic signals

This paper reports how the genetic programming paradigm, in conjunction with pattern recognition principles, can be used to evolve classifiers capable of recognizing epileptic patterns in human electroencephalographic signals. The procedure for feature extraction from the raw signal is detailed, as well as the genetic programming system that properly selects the features and evolves the classifiers. Based on the data sets used, two different epileptic patterns were detected: 3 Hz spike-and-slow-wave-complex (SASWC) and spike-or-sharp-wave (SOSW). After training, classifiers for both patterns were tested with unseen instances, and achieved sensibility = 1.00 and specificity = 0.93 for SASWC patterns, and sensibility = 0.94 and specificity = 0.89 for SOSW patterns. Results are very promising and suggest that the methodology presented can be applied to other pattern recognition tasks in complex signals.     

On reliability of the folded hypercubes

In this paper, we explore the 2-extra connectivity and 2-extra-edge-connectivity of the folded hypercube FQ_n. We show that κ₂(FQ_n) = 3n − 2 for n ⩾ 8; and λ₂(FQ_n) = 3n − 1 for n ⩾ 5. That is, for n ⩾ 8 (resp. n ⩾ 5), at least 3n − 2 vertices (resp. 3n − 1 edges) of FQ_n are removed to get a disconnected graph that contains no isolated vertices (resp. edges). When the folded hypercube is used to model the topological structure of a large-scale parallel processing system, these results can provide more accurate measurements for reliability and fault tolerance of the system.     

Extending multimedia research: How do prerequisite knowledge and reading comprehension affect learning from text and pictures

The present study aimed at extending research on multimedia design principles by investigating their validity as a function of learners’ reading comprehension and scientific literacy. Students (N = 125; age: M = 15.11 years) learned about cell reproduction during their regular Biology lessons in one of six conditions resulting from cross-varying multimedia (text only vs. text plus animations) and text modality (spoken vs. written vs. spoken and written). Recall and transfer were assessed immediately after learning and again 1 week later. Overall, adding animations to text as well as using spoken rather than written text improved only immediate recall; in addition, a multimedia effect for delayed recall was observed for learners with higher levels of scientific literacy. A redundant presentation of text proved harmful especially for delayed performance measures. Reading comprehension did not moderate multimedia design effects. Students with more suitable cognitive prerequisites were better able to maintain performance from the immediate to the delayed tests. Future multimedia research should further investigate the boundary conditions that moderate multimedia effectiveness.     

Human lower extremity joint moment prediction: A wavelet neural network approach

Joint moment is one of the most important factors in human gait analysis. It can be calculated using multi body dynamics but might not be straight forward. This study had two main purposes; firstly, to develop a generic multi-dimensional wavelet neural network (WNN) as a real-time surrogate model to calculate lower extremity joint moments and compare with those determined by multi body dynamics approach, secondly, to compare the calculation accuracy of WNN with feed forward artificial neural network (FFANN) as a traditional intelligent predictive structure in biomechanics.To aim these purposes, data of four patients walked with three different conditions were obtained from the literature. A total of 10 inputs including eight electromyography (EMG) signals and two ground reaction force (GRF) components were determined as the most informative inputs for the WNN based on the mutual information technique. Prediction ability of the network was tested at two different levels of inter-subject generalization. The WNN predictions were validated against outputs from multi body dynamics method in terms of normalized root mean square error (NRMSE (%)) and cross correlation coefficient (ρ).Results showed that WNN can predict joint moments to a high level of accuracy (NRMSE < 10%, ρ > 0.94) compared to FFANN (NRMSE < 16%, ρ > 0.89). A generic WNN could also calculate joint moments much faster and easier than multi body dynamics approach based on GRFs and EMG signals which released the necessity of motion capture. It is therefore indicated that the WNN can be a surrogate model for real-time gait biomechanics evaluation.     

Empirical evaluation of the effects of mixed project data on learning defect predictors

ContextDefect prediction research mostly focus on optimizing the performance of models that are constructed for isolated projects (i.e. within project (WP)) through retrospective analyses. On the other hand, recent studies try to utilize data across projects (i.e. cross project (CP)) for building defect prediction models for new projects. There are no cases where the combination of within and cross (i.e. mixed) project data are used together.ObjectiveOur goal is to investigate the merits of using mixed project data for binary defect prediction. Specifically, we want to check whether it is feasible, in terms of defect detection performance, to use data from other projects for the cases (i) when there is an existing within project history and (ii) when there are limited within project data.MethodWe use data from 73 versions of 41 projects that are publicly available. We simulate the two above-mentioned cases, and compare the performances of naive Bayes classifiers by using within project data vs. mixed project data.ResultsFor the first case, we find that the performance of mixed project predictors significantly improves over full within project predictors (p-value < 0.001), however the effect size is small (Hedges′ g = 0.25). For the second case, we found that mixed project predictors are comparable to full within project predictors, using only 10% of available within project data (p-value = 0.002, g = 0.17).ConclusionWe conclude that the extra effort associated with collecting data from other projects is not feasible in terms of practical performance improvement when there is already an established within project defect predictor using full project history. However, when there is limited project history, e.g. early phases of development, mixed project predictions are justifiable as they perform as good as full within project models.     

Bivariate quality control using two-stage intelligent monitoring scheme

In manufacturing industries, it is well known that process variation is a major source of poor quality products. As such, monitoring and diagnosis of variation is essential towards continuous quality improvement. This becomes more challenging when involving two correlated variables (bivariate), whereby selection of statistical process control (SPC) scheme becomes more critical. Nevertheless, the existing traditional SPC schemes for bivariate quality control (BQC) were mainly designed for rapid detection of unnatural variation with limited capability in avoiding false alarm, that is, imbalanced monitoring performance. Another issue is the difficulty in identifying the source of unnatural variation, that is, lack of diagnosis, especially when dealing with small shifts. In this research, a scheme to address balanced monitoring and accurate diagnosis was investigated. Design consideration involved extensive simulation experiments to select input representation based on raw data and statistical features, artificial neural network recognizer design based on synergistic model, and monitoring–diagnosis approach based on two-stage technique. The study focused on bivariate process for cross correlation function, ρ = 0.1–0.9 and mean shifts, μ = ±0.75–3.00 standard deviations. The proposed two-stage intelligent monitoring scheme (2S-IMS) gave superior performance, namely, average run length, ARL₁ = 3.18–16.75 (for out-of-control process), ARL₀ = 335.01–543.93 (for in-control process) and recognition accuracy, RA = 89.5–98.5%. This scheme was validated in manufacturing of audio video device component. This research has provided a new perspective in realizing balanced monitoring and accurate diagnosis in BQC.     

Modeling of the energies and splitting of the Qx and Qy bands in positional isomers of zinc pyridinoporphyrazines by TDDFT approach: Can TDDFT help distinguishing the structural isomers?

Electronic structures, energies and splitting of the Q_x and Q_y bands for positional isomers of zinc mono-, di-, tri-, and tetra pyridinoporphyrazines as well as parent zinc phthalocyanine were investigated using density functional theory (DFT) and time-dependent (TD) DFT approaches. The influence of the Hartree–Fock exchange on excited state energies and Q_x and Q_y bands splitting were studied using GGA BP86 and hybrid B3LYP and PBE1PBE exchange-correlation functionals. Solvent effects were estimated using the polarized continuum model (PCM) approach and cyclohexane, toluene, or DMSO as solvents. It was found that general trends in the Q_x and Q_y band energies and splitting correlate very well with the available experimental data on pyridinoporphyrazines and follow the trends in HOMO–LUMO and HOMO–LUMO + 1 energy gaps as well as LUMO–LUMO + 1 splitting. TDDFT trends allow estimation of the Q_x and Q_y band energies and splitting in unknown tripyridinoporphyrazines and in individual positional isomers of tetrapyridinoporphyrazines.     

Computer-based analysis of explicit approximations to the implicit Colebrook–White equation in turbulent flow friction factor calculation

The implicit Colebrook–White equation has been widely used to estimate the friction factor for turbulent fluid-flow in rough-pipes. In this paper, the state-of-the-art review for the most currently available explicit alternatives to the Colebrook–White equation, is presented. An extensive comparison test was established on the 20 × 500 grid, for a wide range of relative roughness (ε/D) and Reynolds number (R) values (1 × 10^?6 ? ε/D ? 5 × 10^?2; 4 × 10³ ? R ? 10⁸), covering a large portion of turbulent flow zone in Moody’s diagram. Based on the comprehensive error analysis, the magnitude points in which the maximum absolute and the maximum relative error are occurred at the pair of ε/D and R values, are observed. A limiting case of the most of these approximations provided friction factor estimates that are characterized by a mean absolute error of 5 × 10^?4, a maximum absolute error of 4 × 10^?3 whereas, a mean relative error of 1.3% and a maximum relative error of 5.8%, over the entire range of ε/D and R values, respectively. For practical purposes, the complete results for the maximum and the mean relative errors versus the 20 sets of ε/D value, are also indicated in two comparative figures. The examination results for error properties of these approximations gives one an opportunity to practically evaluate the most accurate formula among of all the previous explicit models; and showing in this way its great flexibility for estimating turbulent flow friction factor. Comparative analysis for the mean relative error profile revealed, the classification for the best-fitted six equations examined was in a good agreement with those of the best model selection criterion claimed in the recent literature, for all performed simulations.     

Bounds for the Roots of Lacunary Polynomials

A polynomial P(X)  = X^d + a_d − 1X^d − 1 + ⋯ is called lacunary when a_d − 1 =  0. We give bounds for the roots of such polynomials with complex coefficients. These bounds are much smaller than for general polynomials.     

Hexagon-based all-quadrilateral mesh generation with guaranteed angle bounds

In this paper, we present a novel hexagon-based mesh generation method which creates all-quadrilateral (all-quad) meshes with guaranteed angle bounds and feature preservation for arbitrary planar domains. Given any planar curves, an adaptive hexagon-tree structure is constructed by using the curvature of the boundaries and narrow regions. Then a buffer zone and a hexagonal core mesh are created by removing elements outside or around the boundary. To guarantee the mesh quality, boundary edges of the core mesh are adjusted to improve their formed angles facing the boundary, and two layers of quad elements are inserted in the buffer zone. For any curve with sharp features, a corresponding smooth curve is firstly constructed and meshed, and then another layer of elements is inserted to match the smooth curve with the original one. It is proved that for any planar smooth curve all the element angles are within [60° ? ε, 120° + ε] (ε ? 5°). We also prove that the scaled Jacobians defined by two edge vectors are in the range of [sin (60° ? ε),  sin 90°], or [0.82, 1.0]. The same angle range can be guaranteed for curves with sharp features, with the exception of small angles in the input curve. Furthermore, an approach is introduced to match the generated interior and exterior meshes with a relaxed angle range, [30°, 150°]. We have applied our algorithm to a set of complicated geometries, including the China map, the Lake Superior map, and a three-component air foil with sharp features. In addition, all the elements in the final mesh are grouped into five types, and most elements only need a few flops to construct the stiffness matrix for finite element analysis. This will significantly reduce the computational time and the required memory during the stiffness matrix construction.     

Performance of radial basis and LM-feed forward artificial neural networks for predicting daily watershed runoff

This study investigated the effects of upstream stations’ flow records on the performance of artificial neural network (ANN) models for predicting daily watershed runoff. As a comparison, a multiple linear regression (MLR) analysis was also examined using various statistical indices. Five streamflow measuring stations on the Cahaba River, Alabama, were selected as case studies. Two different ANN models, multi layer feed forward neural network using Levenberg–Marquardt learning algorithm (LMFF) and radial basis function (RBF), were introduced in this paper. These models were then used to forecast one day ahead streamflows. The correlation analysis was applied for determining the architecture of each ANN model in terms of input variables. Several statistical criteria (RMSE, MAE and coefficient of correlation) were used to check the model accuracy in comparison with the observed data by means of K-fold cross validation method. Additionally, residual analysis was applied for the model results. The comparison results revealed that using upstream records could significantly increase the accuracy of ANN and MLR models in predicting daily stream flows (by around 30%). The comparison of the prediction accuracy of both ANN models (LMFF and RBF) and linear regression method indicated that the ANN approaches were more accurate than the MLR in predicting streamflow dynamics. The LMFF model was able to improve the average of root mean square error (RMSE_ave) and average of mean absolute percentage error (MAPE_ave) values of the multiple linear regression forecasts by about 18% and 21%, respectively. In spite of the fact that the RBF model acted better for predicting the highest range of flow rate (flood events, RMSE_ave/RBF = 26.8 m³/s vs. RMSE_ave/LMFF = 40.2 m³/s), in general, the results suggested that the LMFF method was somehow superior to the RBF method in predicting watershed runoff (RMSE/LMFF = 18.8 m³/s vs. RMSE/RBF = 19.2 m³/s). Eventually, statistical differences between measured and predicted medians were evaluated using Mann-Whitney test, and differences in variances were evaluated using the Levene's test.     

A telemedicine tool to detect pulmonary pathology using computerized pulmonary acoustic signal analysis

BackgroundDetection and monitoring of respiratory related illness is an important aspect in pulmonary medicine. Acoustic signals extracted from the human body are considered in detection of respiratory pathology accurately.ObjectivesThe aim of this study is to develop a prototype telemedicine tool to detect respiratory pathology using computerized respiratory sound analysis.MethodsAround 120 subjects (40 normal, 40 continuous lung sounds (20 wheeze and 20 rhonchi)) and 40 discontinuous lung sounds (20 fine crackles and 20 coarse crackles) were included in this study. The respiratory sounds were segmented into respiratory cycles using fuzzy inference system and then S-transform was applied to these respiratory cycles. From the S-transform matrix, statistical features were extracted. The extracted features were statistically significant with p < 0.05. To classify the respiratory pathology KNN, SVM and ELM classifiers were implemented using the statistical features obtained from of the data.ResultsThe validation showed that the classification rate for training for ELM classifier with RBF kernel was high compared to the SVM and KNN classifiers. The time taken for training the classifier was also less in ELM compared to SVM and KNN classifiers. The overall mean classification rate for ELM classifier was 98.52%.ConclusionThe telemedicine software tool was developed using the ELM classifier. The telemedicine tool has performed extraordinary well in detecting the respiratory pathology and it is well validated.     

A simple semi-analytical model for remote estimation of chlorophyll-a in turbid waters: Validation

Accurate assessment of phytoplankton chlorophyll-a (chla) concentrations in turbid waters by means of remote sensing is challenging due to the optical complexity of case 2 waters. We have applied a recently developed model of the form [R_rs^? 1(λ₁) ? R_rs^? 1(λ₂)] × R_rs(λ₃) where R_rs(λ_i) is the remote-sensing reflectance at the wavelength λ_i, for the estimation of chla concentrations in turbid waters. The objectives of this paper are (a) to validate the three-band model as well as its special case, the two-band model R_rs^? 1(λ₁) × R_rs(λ₃), using datasets collected over a considerable range of optical properties, trophic status, and geographical locations in turbid lakes, reservoirs, estuaries, and coastal waters, and (b) to evaluate the extent to which the three-band model could be applied to the Medium Resolution Imaging Spectrometer (MERIS) and two-band model could be applied to the Moderate Resolution Imaging Spectroradiometer (MODIS) to estimate chla in turbid waters.The three-band model was calibrated and validated using three MERIS spectral bands (660–670 nm, 703.75–713.75 nm, and 750?757.5 nm), and the 2-band model was tested using two MODIS spectral bands (λ₁ = 662–672, λ₃ = 743–753 nm). We assessed the accuracy of chla prediction in four independent datasets without re-parameterization (adjustment of the coefficients) after initial calibration elsewhere. Although the validation data set contained widely variable chla (1.2 to 236 mg m^? 3), Secchi disk depth (0.18 to 4.1 m), and turbidity (1.3 to 78 NTU), chla predicted by the three-band algorithm was strongly correlated with observed chla (r² > 0.96), with a precision of 32% and average bias across data sets of ? 4.9% to 11%. Chla predicted by the two-band algorithm was also closely correlated with observed chla (r² > 0.92); however, the precision declined to 57%, and average bias across the data sets was 18% to 50.3%. These findings imply that, provided that an atmospheric correction scheme for the red and NIR bands is available, the extensive database of MERIS and MODIS imagery could be used for quantitative monitoring of chla in turbid waters.     

Computational assessment of several hydrogen-free high energy compounds

Tetrazino-tetrazine-tetraoxide (TTTO) is an attractive high energy compound, but unfortunately, it is not yet experimentally synthesized so far. Isomerization of TTTO leads to its five isomers, bond-separation energies were empolyed to compare the global stability of six compounds, it is found that isomer 1 has the highest bond-separation energy (1204.6 kJ/mol), compared with TTTO (1151.2 kJ/mol); thermodynamic properties of six compounds were theoretically calculated, including standard formation enthalpies (solid and gaseous), standard fusion enthalpies, standard vaporation enthalpies, standard sublimation enthalpies, lattice energies and normal melting points, normal boiling points; their detonation performances were also computed, including detonation heat (Q, cal/g), detonation velocity (D, km/s), detonation pressure (P, GPa) and impact sensitivity (h₅₀, cm), compared with TTTO (Q = 1311.01 J/g, D = 9.228 km/s, P = 40.556 GPa, h₅₀ = 12.7 cm), isomer 5 exhibites better detonation performances (Q = 1523.74 J/g, D = 9.389 km/s, P = 41.329 GPa, h₅₀ =  28.4 cm).     

Dielectric and piezoelectric properties of [0 0 1] and [0 1 1]-poled relaxor ferroelectric PZN–PT and PMN–PT single crystals

The properties of PZN–PT and PMN–PT single crystals of varying compositions and orientations have been investigated. Among the various compositions studied, [0 0 1]-optimally poled PZN-(6–7)%PT and PMN-30%PT exhibit superior dielectric and piezoelectric properties, with K^T ≈ 6800–8000, d₃₃ ≈ 2800 pC/N, d₃₁ ≈ −(1200–1800) pC/N for PZN-(6–7)%PT; and K^T = 7500–9000, d₃₃ = 2200–2500 pC/N and d₃₁ = −(1100–1400) pC/N for PMN-30%PT. These two compositions are also fairly resistant to over-poling. The [0 0 1]-poled electromechanical coupling factors (k₃₃, k₃₁ and k_t) are relatively insensitive to crystal composition. [0 1 1]-optimally poled PZN-7%PT single crystal also exhibits extremely high d₃₁ values of up to −4000 pC/N with k₃₁ ≈ 0.90–0.96. While [0 1 1]-poled PZN-7%PT single crystal becomes over-poled with much degraded properties when poled at and above 0.6 kV/mm, PZN-6%PT crystal shows no signs of over-poling even when poled to 2.0 kV/mm. The presence of a certain amount (i.e., 10–15%) of orthorhombic phase in a rhombohedral matrix has been found to be responsible for the superior transverse piezoelectric properties of [0 1 1]-optimally poled PZN-(6–7)%PT. The present work shows that flux-grown PZN–PT crystals exhibit superior and consistent properties and improved over-poling resistance to flux-grown PMN–PT crystals and that, for or a given crystal composition, flux-grown PMN–PT crystals exhibit superior over-poling resistance to their melt-grown counterparts.     

Thick film sensors based on laccases from different sources immobilized in polyaniline matrix

Thick film electrode based biosensors containing Trametes versicolor (TvL) and Aspergillus niger (AnL) laccases and Agaricus bisporus tissues (AbT) were developed for the determination of phenolic compounds and the measurement was based on oxygen consumption in relation to analyte oxidation. The electrodeposited organic polymer; polyaniline was used as a matrix for the immobilization in the preparation of thick film sensors. The systems were calibrated for different phenolic substances. A linearity was obtained in concentration range between 0.4 and 6.0 μM phenol, 0.2 and 1.0 μM catechol, 2.0 and 20.0 μM l-DOPA for TvL based biosensor; for AnL based enzyme electrode 0.4 and 4.0 μM phenol, 0.4 and 15 μM catechol, 0.4 and 6.0 μM l-DOPA; for AbT electrode 1.0 and 10 μM phenol, 0.4 and 1.6 μM catechol, 1.0 and 10 μM l-DOPA, respectively, in the response time of 300 s. Furthermore, as well as sample application and accuracy, optimum pH, temperature and thermal stabilities of the proposed systems were also detected.