• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.10
• /
• pp.1422-1432
• /
• 1993

Nonlinear mapping function of the HCNN( Hidden Control Neural Network ) can change over time to model the temporal variability of a speech signal by combining the nonlinear prediction of conventional neural networks with the segmentation capability of HMM. We have two things in this paper. first, we showed that the performance of the HCNN is better than that of HMM. Second, the HCNN with its prediction error measure given by weighted distance is proposed to use suitable distance measure for the HCNN, and then we showed that the superiority of the proposed system for speaker-independent speech recognition tasks. Weighted distance considers the differences between the variances of each component of the feature vector extraced from the speech data. Speaker-independent Korean digit recognition experiment showed that the recognition rate of 95%was obtained for the HCNN with Euclidean distance. This result is 1.28% higher than HMM, and shows that the HCNN which models the dynamical system is superior to HMM which is based on the statistical restrictions. And we obtained 97.35% for the HCNN with weighted distance, which is 2.35% better than the HCNN with Euclidean distance. The reason why the HCNN with weighted distance shows better performance is as follows : it reduces the variations of the recognition error rate over different speakers by increasing the recognition rate for the speakers who have many misclassified utterances. So we can conclude that the HCNN with weighted distance is more suit-able for speaker-independent speech recognition tasks.

• The Journal of Engineering Geology
• /
• v.14 no.1
• /
• pp.47-60
• /
• 2004

The permeability coefficients were calculated by the homogenization analysis method with sufficient consideration of fracture geometry dependent on aperture change. According to the results of aperture measurements using a confocal laser scanning microscope, apertures on each measuring point display different magnitudes, indicating that fracture walls can not be assumed as parallel feature. After construction of fracture model based on the aperture values measured on each pressure level, the homogenization analysis was conducted to compute permeability coefficients. The calculated permeability coefficients distribute in the ranges of $10^{-1}~10^{-3}cm/sec$. Most of the specimens show decreasing permeability coefficients with the increase of the applied pressure. However, the decreasing rates of permeability coefficients do not show a constant trend on each pressure level. This phenomenon is well matched to the observation results of Chae et al. (2003). It proves that aperture change strongly influences on permeability characteristics. Three sections of each specimen have all different values of permeability coefficient. It suggests that the variation of permeability coefficient depends sensitively on aperture magnitudes and characteristics of fracture geometry. It is very important to consider accurate fracture geometries for analysis of permeability characteristics in rock fractures bearing different aperture distribution. Therefore, it needs to consider sufficiently the fracture geometries for calculating the permeability coefficients of fractures.

• Food Engineering Progress
• /
• v.14 no.1
• /
• pp.49-53
• /
• 2010

Surface plasmon resonance (SPR) which is utilized in thin film refractometry-based sensors has been concerned on measurement of physical, chemical and biological quantities because of its high sensitivity and label-free feature. In this paper, an application of SPR to detection of alcohol content in wine and liquor was investigated. The result showed that SPR sensor had high potential to evaluate alcohol content. Nevertheless, food industry may need SPR sensor with higher sensitivity. Herein, we introduced a nano-technique into fabrication of SPR chip to enhance SPR sensitivity. Using Langmuir-Blodgett (LB) method, gold film with nano-structured surface was devised. In order to make a new SPR chip, firstly, a single layer of nano-scaled silica particles adhered to plain surface of gold film. Thereafter, gold was deposited on the template by an e-beam evaporator. Finally, the nano-structured surface with basin-like shape was obtained after removing the silica particles by sonication. In this study, two types of silica particles, or 130 nm and 300 nm, were used as template beads and sensitivity of the new SPR chip was tested with ethanol solution, respectively. Applying the new developed SPR sensor to a model food of alcoholic beverage, the sensitivity showed improvement of 95% over the conventional one.

• Asian-Australasian Journal of Animal Sciences
• /
• v.32 no.7
• /
• pp.1015-1026
• /
• 2019

Objective: The objective of this study was to evaluate three different degrees of white striping (WS) addressing their automatic assessment and customer acceptance. The WS classification was performed based on a computer vision system (CVS), exploring different machine learning (ML) algorithms and the most important image features. Moreover, it was verified by consumer acceptance and purchase intent. Methods: The samples for image analysis were classified by trained specialists, according to severity degrees regarding visual and firmness aspects. Samples were obtained with a digital camera, and 25 features were extracted from these images. ML algorithms were applied aiming to induce a model capable of classifying the samples into three severity degrees. In addition, two sensory analyses were performed: 75 samples properly grilled were used for the first sensory test, and 9 photos for the second. All tests were performed using a 10-cm hybrid hedonic scale (acceptance test) and a 5-point scale (purchase intention). Results: The information gain metric ranked 13 attributes. However, just one type of image feature was not enough to describe the phenomenon. The classification models support vector machine, fuzzy-W, and random forest showed the best results with similar general accuracy (86.4%). The worst performance was obtained by multilayer perceptron (70.9%) with the high error rate in normal (NORM) sample predictions. The sensory analysis of acceptance verified that WS myopathy negatively affects the texture of the broiler breast fillets when grilled and the appearance attribute of the raw samples, which influenced the purchase intention scores of raw samples. Conclusion: The proposed system has proved to be adequate (fast and accurate) for the classification of WS samples. The sensory analysis of acceptance showed that WS myopathy negatively affects the tenderness of the broiler breast fillets when grilled, while the appearance attribute of the raw samples eventually influenced purchase intentions.

• The Korean Journal of Applied Statistics
• /
• v.33 no.6
• /
• pp.777-789
• /
• 2020

COVID-19 has been rapidly spread world wide since late December 2019. In this paper, our interest is to estimate distribution of incubation time defined as period between infection of virus and the onset. Due to the limit of accessibility and asymptomatic feature of COVID-19 virus, the exact infection and onset time are not always observable. For estimation of incubation time, interval censoring technique is implemented. Furthermore, a competing risk model is applied to estimate the case fatality and cure fraction. Based on the result, the mean incubation time is about 5.4 days and the fatality rate is higher for older and male patient and the cure rate is higher at younger,female and asymptomatic patient.

• Journal of the Korean Society of Radiology
• /
• v.15 no.4
• /
• pp.547-554
• /
• 2021

In order to quickly and accurately diagnose pneumonia on a chest X-ray image, different batch sizes of 4, 8, 16, and 32 were applied to the same Xception deep learning model, and modeling was performed 3 times, respectively. As a result of the performance evaluation of deep learning modeling, in the case of modeling to which batch size 32 was applied, the results of accuracy, loss function value, mean square error, and learning time per epoch showed the best results. And in the accuracy evaluation of the Test Metric, the modeling applied with batch size 8 showed the best results, and the precision evaluation showed excellent results in all batch sizes. In the recall evaluation, modeling applied with batch size 16 showed the best results, and for F1-score, modeling applied with batch size 16 showed the best results. And the AUC score evaluation was the same for all batch sizes. Based on these results, deep learning modeling with batch size 32 showed high accuracy, stable artificial neural network learning, and excellent speed. It is thought that accurate and rapid lesion detection will be possible if a batch size of 32 is applied in an automatic diagnosis study for feature extraction and classification of pneumonia in chest X-ray images using deep learning in the future.

• Journal of the Korea Convergence Society
• /
• v.12 no.7
• /
• pp.273-282
• /
• 2021

Nursing home plays a role in providing elderly care in the context of China's rapid population aging, but little understanding of the efficiency of the nursing homes. In this paper, we investigated the efficiency in nursing homes using Data Envelopment Analysis (DEA) and Malmquist index (MPI) for the modeling of the number of nursing home beds, fixed assets, and medical personnel as input variables, and the number of elderly people of self-care, the number of elderly people of partial self-care, the number of bed-ridden elderly people and the income of nursing homes as output variables. Stratification analysis showed that the top two provinces in the DEA-CCR yield were Beijing and Shanghai in the five-year survey period. Four provinces (Beijing, Jiangsu, Shandong, and Shanghai) scored 1.00 in terms of DEA-BCC yield. The MPI analysis showed that Hainan ranked the highest five-year average in the included provinces. In terms of resource utilization, internal management, operation scale, and other aspects, the nursing homes in the provinces with high-efficiency evaluation results show high efficiency and technological progress, whereas the areas with low-efficiency evaluation showed a feature of the improving technical efficiency.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.135-144
• /
• 2021

Recently, data-driven decision-making technology has become a key technology leading the data industry, and machine learning technology for this requires high-quality training datasets. However, real-world data contains missing values for various reasons, which degrades the performance of prediction models learned from the poor training data. Therefore, in order to build a high-performance model from real-world datasets, many studies on automatically imputing missing values in initial training data have been actively conducted. Many of conventional machine learning-based imputation techniques for handling missing data involve very time-consuming and cumbersome work because they are applied only to numeric type of columns or create individual predictive models for each columns. Therefore, this paper proposes a new data imputation technique called 'Denoising Self-Attention Network (DSAN)', which can be applied to mixed-type dataset containing both numerical and categorical columns. DSAN can learn robust feature expression vectors by combining self-attention and denoising techniques, and can automatically interpolate multiple missing variables in parallel through multi-task learning. To verify the validity of the proposed technique, data imputation experiments has been performed after arbitrarily generating missing values for several mixed-type training data. Then we show the validity of the proposed technique by comparing the performance of the binary classification models trained on imputed data together with the errors between the original and imputed values.

• Asia Marketing Journal
• /
• v.13 no.4
• /
• pp.253-264
• /
• 2012

A growing number of Korean companies are trying to expand their business area into global market due to saturation in the Korean domestic market. Home shopping industry arriving on mature stage is faced with less growth recently. CJ O Shopping which is a top ranked home shopping company in Korea, has been showing meaningful performances by earlier moving to global market with thorough preparations. CJ O Shopping's global marketing strategy focused on asian countries including China, India, Vietnam, and Japan is going successfully, which enables top ranked on-line retailing company in asia as well as in Korea. CJ O Shopping effectively penetrated into overseas market with both core competence based on Korean home shopping model and rigorous preliminary study on target market. Especially shoppertainment (Shopping+Entertainment) that is unique feature of globally competitive Korean home shopping created huge differentiations in target market. Also choosing the influential local partner, sharing the business goals, and building the joint venture could make stable operations, thereby easily earning of well-established awareness from target consumers. A step ahead entry of competitors and intensive localization of CJ O Shopping's core competence for arriving safe in target market were additional key factors for global marketing success. We can extract above key factors for success as implications of case study on CJ O Shopping's global marketing, and expect those factors to be spread into lots of Korean companies and utilized as successful strategies for global marketing.

• Journal of the Korea Society for Simulation
• /
• v.18 no.3
• /
• pp.103-112
• /
• 2009

In a nuclear power plant (NPP), periodic sensor calibrations are required to assure sensors are operating correctly. However, only a few faulty sensors are found to be calibrated. For the safe operation of an NPP and the reduction of unnecessary calibration, on-line calibration monitoring is needed. In this paper, principal component-based Auto-Associative support vector regression (PCSVR) was proposed for the sensor signal validation of the NPP. It utilizes the attractive merits of principal component analysis (PCA) for extracting predominant feature vectors and AASVR because it easily represents complicated processes that are difficult to model with analytical and mechanistic models. With the use of real plant startup data from the Kori Nuclear Power Plant Unit 3, SVR hyperparameters were optimized by the response surface methodology (RSM). Moreover the statistical techniques are integrated with PCSVR for the failure detection. The residuals between the estimated signals and the measured signals are tested by the Shewhart Control Chart, Exponentially Weighted Moving Average (EWMA), Cumulative Sum (CUSUM) and generalized likelihood ratio test (GLRT) to detect whether the sensors are failed or not. This study shows the GLRT can be a candidate for the detection of sensor drift.