Search

Research on the detection of solar events has been conducted over many years. Recently, deep learning and data-driven approaches have been applied to solar event recognition. In this study, we present solar event detection using deep-learning-based object detection methods for real-time space weather monitoring. First, we construct a new object detection dataset using imaging data obtained by the Solar Dynamics Observatory with bounding boxes as labels for three representative features: coronal holes, sunspots, and prominences. Second, we train two representative object detection models: the Single Shot MultiBox Detector (SSD) and the Faster Region-based Convolutional Neural Network (R-CNN) using the new dataset. The results show that both models perform similarly well for coronal hole and sunspot detection. For prominence detection, the SSD and Faster R-CNN exhibited relatively low performance. This study demonstrates that deep-learning-based object detection can successfully detect multiple types of solar events, and it may be extended to detect other solar events. In addition, we provide the dataset for further achievements of object detection studies in solar physics.

Introduction: Globally, more people are spending time on the Internet and gaming since the outbreak of the Coronavirus Disease 2019 (COVID-19). Consequently, concerns about developing behavioral addiction of adolescents have been raised. Such risk could be greater for adolescents in South Korea where the majority of adolescents have access to the Internet and own a smartphone. In fact, statistics indicate that Korean youths are spending significantly more time on the Internet and gaming during the COVID-19 pandemic. Previous studies on the patterns of time spent on the Internet and Internet gaming show inconsistent results. The aim of this study is to investigate the latent profiles of the Internet and Internet game usage among adolescents in South Korea.Method: Data from a national survey on elementary and middle school students across South Korea were used. The sample consists of 3,149 respondents, and 2,984 responses were analyzed after removing missing responses. Latent profile analysis was performed to investigate the number of latent profiles for the Internet and Internet game usage time. To validate the profiles, differences in problematic gaming behavior, sex, and neuroticism were examined.Results: Seven profiles were found: Casual User, Moderate User, Smartphone User, Internet User, PC Internet Gamer, Heavy User, and Excessive User. Validation of the profiles indicated differences in problematic gaming behavior, sex, and neuroticism among selected profiles.Conclusion: This study presented different profiles of the Internet and Internet game usage among adolescents in South Korea. Profiles with higher game usage time scored higher in problematic game use compared to other profiles. Males were more likely to be in the profiles with high gaming time, and females were more likely to be in Internet and Smartphone User profiles. The results indicate that Internet and Internet gaming usage patterns could be classified by the type of device used and the content of the Internet.

The purpose of this study was to analyse online newspaper articles on students with intellectual disabilities (IDs) in order to identify related social phenomena to derive implications for inclusive education. Such study has traditionally practised through content analysis and/or discourse analysis manually, which is prone to subjective interpretation. Thus, this study implemented automated analysis to objectively select and interpret a big data. A total of 8,890 online newspaper articles that were published from 1990 to April 2019 were collected through automated parsing. The entire period and decade-phase based keyword and keyword network analysis were practised in order to determine how the social perceptions and related issues had changed over time. The results indicated that there was a rapid growth in scope of articles on students with IDs over the past 30 years. The attention of media gradually expanded from special education to improving quality of lives of students with IDs and their families. Moreover, online newspaper articles seemed to focus on social controversies and incidents such as sexual assaults that are related to students with IDs. Based on the results, ways to support inclusive education as well as social inclusion of students with IDs were discussed.

One-class support vector machine (OCSVM) has demonstrated superior performance in one-class classification problems. However, its training is impractical for large-scale datasets owing to high computational complexity with respect to the number of training instances. In this study, we propose an approximate training method based on the concept of expected margin to obtain a model identical to full training with reduced computational burden. The proposed method selects prospective support vectors using multiple OCSVM models trained on small bootstrap samples of an original dataset. The final OCSVM model is trained using only the selected instances. The proposed method is not only simple and straightforward but also considerably effective in improving the training efficiency of OCSVM. Preliminary experiments are conducted on large-scale benchmark datasets to examine the effectiveness of the proposed method in terms of approximation performance and computational efficiency.

Since the outbreak of the COVID-19 pandemic, the heightened risk of school closures and mental disorders has made adolescents particularly vulnerable to developing internet gaming disorder (IGD). There have been reports of increased time spent playing games on the internet among adolescents during the pandemic, and the risk of developing IGD may be higher for adolescents in South Korea as the majority of them play games on the internet. However, to the best of our knowledge, no studies have examined the impact of the pandemic on adolescents’ internet gaming behavior in South Korea. This study aimed to explore the different profiles of addictive internet gaming behavior among adolescents before and after the outbreak of the COVID-19 pandemic and examine how the pandemic influenced addictive internet gaming usage and time spent playing games on the internet. Nationally representative survey data from the Ministry of Gender Equality and Family with 3040 and 2906 responses from 2018 and 2020, respectively, were analyzed. Using seven factors of a maladaptive gaming usage scale (tolerance, withdrawal, excessive usage, control impairment, compulsive usage, neglecting daily activity, and gaming despite negative consequence), a four-profile model was selected in both 2018 and 2020 for latent profile analysis: ‘casual’ gamer, ‘moderate’ gamer, ‘potential-risk’ gamer and ‘addictive’ gamer. The results from the two-way ANCOVA showed significant interaction between the cohorts (2018 cohort vs. 2020 cohort) and the four profiles on addictive internet gaming usage (F = 119.747, p &lt, 0.001, η2 = 0.05), including time spent playing internet games on a PC (F = 22.893, p &lt, 0.001, η2 = 0.013), and time spent playing games on a mobile phone (F = 3.245, p &lt, 0.05, η2 = 0.02). The results indicated that the increase of addictive internet gaming usage and gameplay time differed by profile. The results imply that the increase in gameplay time was higher for profiles with higher scores in addictive internet gaming usage for internet games played on a PC while the relationship was not obvious for games played on a mobile phone. Despite the statistical significance, there was only 1.2% to 4.9% of mean difference in addictive internet gaming usage between the 2018 and 2020 cohorts, which implies little clinical significance. While adolescents of the four profiles showed no significant signs of increased addictive internet gaming usage, the addictive gamer profile demonstrated a significant increase in game time after COVID-19.

The present study is designed to examine college professors’ perception of classroom disability support services. A survey questionnaire was developed to measure college professors’ perception of three areas including: 1) expanding higher education opportunities for students with disabilities, 2) types of classroom disability support services, and 3) barriers in providing services, and implemented with eighty-two college professors. Overall, results indicate that college professors are permissive in providing disability-related classroom support, although they showed differences in terms of the type and scope of service depending on the nature of the discipline. To improve college disability support service in South Korea, the present study suggests to increase educational opportunities such as booklets, guidelines, and training workshop, specifically designed for college professors who are not familiar with classroom disability support services.

Opioid-induced respiratory depression (OIRD) causes death following an opioid overdose, yet the neurobiological mechanisms of this process are not well understood. Here, we show that neurons within the lateral parabrachial nucleus that express the μ-opioid receptor (PBLOprm1 neurons) are involved in OIRD pathogenesis. PBLOprm1 neuronal activity is tightly correlated with respiratory rate, and this correlation is abolished following morphine injection. Chemogenetic inactivation of PBLOprm1 neurons mimics OIRD in mice, whereas their chemogenetic activation following morphine injection rescues respiratory rhythms to baseline levels. We identified several excitatory G-protein coupled receptors expressed by PBLOprm1 neurons and show that agonists for these receptors restore breathing rates in mice experiencing OIRD. Thus, PBLOprm1 neurons are critical for OIRD pathogenesis, providing a promising therapeutic target for treating OIRD in patients.

Virtual metrology (VM) has been successfully applied to semiconductor manufacturing as an efficient way of achieving wafer-to-wafer quality control. VM involves the estimation of metrology variables of wafer inspection using a prediction model trained with process parameters and measurements prior to the actual implementation of metrology. VM modeling should incorporate a number of process parameters and measurements collected from each piece of process equipment, which results in a greater number of input variables. Therefore, it is necessary to resolve the problem of high dimensionality through feature selection. A suitable feature selection method for VM modeling should effectively address the high dimensionality by lowering the computational cost, while also achieving high prediction accuracy as an essential requirement for the practical deployment of VM. In this paper, a feature selection method based on random forward search is proposed for efficient VM modeling. This method selects relevant variables sequentially from disjoint random subsets of candidate variables by incorporating randomness. Our preliminary experimental results obtained with real-world semiconductor manufacturing data demonstrate that the proposed feature selection method achieves comparable prediction accuracy yet has the advantages of being computationally more efficient, thus merits further investigation.

A new semi-supervised support vector regression method is proposed.Label distribution is estimated by probabilistic local reconstruction algorithm.Different oversampling rate is used based on uncertainty information.Expected margin based pattern selection is used to reduce the training complexity.The proposed method improves the prediction performance with lower time complexity. Dataset size continues to increase and data are being collected from numerous applications. Because collecting labeled data is expensive and time consuming, the amount of unlabeled data is increasing. Semi-supervised learning (SSL) has been proposed to improve conventional supervised learning methods by training from both unlabeled and labeled data. In contrast to classification problems, the estimation of labels for unlabeled data presents added uncertainty for regression problems. In this paper, a semi-supervised support vector regression (SS-SVR) method based on self-training is proposed. The proposed method addresses the uncertainty of the estimated labels for unlabeled data. To measure labeling uncertainty, the label distribution of the unlabeled data is estimated with two probabilistic local reconstruction (PLR) models. Then, the training data are generated by oversampling from the unlabeled data and their estimated label distribution. The sampling rate is different based on uncertainty. Finally, expected margin-based pattern selection (EMPS) is employed to reduce training complexity. We verify the proposed method with 30 regression datasets and a real-world problem: virtual metrology (VM) in semiconductor manufacturing. The experiment results show that the proposed method improves the accuracy by 8% compared with conventional supervised SVR, and the training time for the proposed method is 20% shorter than that of the benchmark methods.

As modern-day adolescents use the Internet on both personal computer (PC) and smartphone, this study examined the phenomenon of problematic internet use by taking account of Internet usage on both PC and smartphone together, based on the theoretical framework of substitution/complementarity of media use. For this, latent profile analysis, nonlinear canonical correlation analysis, and logistic/probit regression analyses were performed on 653 Korean adolescents. Latent profile analysis identified six classes of distinct problematic internet use patterns. In brief, two latent classes showed substituting patterns, two other classes showed complementing patterns, and the last two showed neither. According to nonlinear canonical correlation analysis, classification by latent profile analysis was mainly associated with individual variables such as 'PC game,' 'instant messaging,' 'gender,' and 'decreased PC usage time.' Further, logistic/probit regression analyses revealed that male adolescents were more likely to be included in the complementation class, because they played PC games more than female adolescents. Implications and limitations of the study are discussed. Problematic internet use on both PC and smartphone were examined together.Latent profile analysis identified six classes of problematic internet use patterns.More males were included in the complementation class because of PC games.

Support Vector Machines (SVMs) are amongst the most powerful classification algorithms in machine learning and data mining. However, SVMs are limited by high training complexity when training with large datasets. Pattern selection methods have been proposed to reduce the training complexity by selecting a smaller subset of important patterns among all training patterns. In this paper, we propose a new pattern selection method called Expected Margin–based Pattern Selection (EMPS), which selects patterns based on an estimated margin for SVM classifiers. With the estimated margin, EMPS selects patterns that are likely to become support vectors located on the margin boundary and inside the margin region; however, other patterns including noise support vectors are discarded. The experimental results involving 15 benchmark datasets and one real–world semiconductor manufacturing dataset showed that EMPS exhibits excellent performance and stability.

Internet addiction has become a serious behavioral health problem in Asia. However, there are no up-to-date country comparisons. The Asian Adolescent Risk Behavior Survey (AARBS) screens and compares the prevalence of Internet behaviors and addiction in adolescents in six Asian countries. A total of 5,366 adolescents aged 12-18 years were recruited from six Asian countries: China, Hong Kong, Japan, South Korea, Malaysia, and the Philippines. Participants completed a structured questionnaire on their Internet use in the 2012-2013 school year. Internet addiction was assessed using the Internet Addiction Test (IAT) and the Revised Chen Internet Addiction Scale (CIAS-R). The variations in Internet behaviors and addiction across countries were examined. The overall prevalence of smartphone ownership is 62%, ranging from 41% in China to 84% in South Korea. Moreover, participation in online gaming ranges from 11% in China to 39% in Japan. Hong Kong has the highest number of adolescents reporting daily or above Internet use (68%). Internet addiction is highest in the Philippines, according to both the IAT (5%) and the CIAS-R (21%). Internet addictive behavior is common among adolescents in Asian countries. Problematic Internet use is prevalent and characterized by risky cyberbehaviors.

Objectives: It is necessary to apply quantitative structure activity relationship(QSAR) for the various chemicals with insufficient toxicity data that are used in the workplace, based on the precautionary principle. This study aims to find application plan of QSAR software tool for predicting health hazards such as genetic toxicity, and carcinogenicity for some chemicals used in the electronics industries. Methods: Toxicity prediction of 21 chemicals such as 5-aminotetrazole, ethyl lactate, digallium trioxide, etc. used in electronics industries was assessed by Toxicity Prediction by Komputer Assisted Technology(TOPKAT). In order to identify the suitability and reliability of carcinogenicity prediction, 25 chemicals such as 4-aminobiphenyl, ethylene oxide, etc. which are classified as Group 1 carcinogens by the International Agency for Research on Cancer(IARC) were selected. Results: Among 21 chemicals, we obtained prediction results for 5 carcinogens, 8 non-carcinogens and 8 unpredictability chemicals. On the other hand, the carcinogenic potential of 5 carcinogens was found to be low by relevant research testing data and Oncologic TM tool. Seven of the 25 carcinogens(IARC Group 1) were wrongly predicted as non-carcinogens(false negative rate: 36.8%). We confirmed that the prediction error could be improved by combining genetic toxicity information such as mutagenicity. Conclusions: Some compounds, including inorganic chemicals and polymers, were still limited for applying toxicity prediction program. Carcinogenicity prediction may be further improved by conducting cross-validation of various toxicity prediction programs, or application of the theoretical molecular descriptors.

Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.

Machine learning has been applied successfully for faulty wafer detection tasks in semiconductor manufacturing. For the tasks, prediction models are built with prior data to predict the quality of future wafers as a function of their precedent process parameters and measurements. In real-world problems, it is common for the data to have a portion of input variables that are irrelevant to the prediction of an output variable. The inclusion of many irrelevant variables negatively affects the performance of prediction models. Typically, prediction models learned by different learning algorithms exhibit different sensitivities with regard to irrelevant variables. Algorithms with low sensitivities are preferred as a first trial for building prediction models, whereas a variable selection procedure is necessarily considered for highly sensitive algorithms. In this study, we investigate the effect of irrelevant variables on three well-known representative learning algorithms that can be applied to both classification and regression tasks: artificial neural network, decision tree (DT), and k-nearest neighbors (k-NN). We analyze the characteristics of these learning algorithms in the presence of irrelevant variables with different model complexity settings. An empirical analysis is performed using real-world datasets collected from a semiconductor manufacturer to examine how the number of irrelevant variables affects the behavior of prediction models trained with different learning algorithms and model complexity settings. The results indicate that the prediction accuracy of k-NN is highly degraded, whereas DT demonstrates the highest robustness in the presence of many irrelevant variables. In addition, a higher model complexity of learning algorithms leads to a higher sensitivity to irrelevant variables.

Objectives This study developed an information and communication technology (ICT)-based comprehensive health and social-needs assessment (CHSNA) system based on the International Classification of Functioning, Disability, and Health (ICF) with the aim of enhancing person-centered community care for community residents and supporting healthcare professionals and social workers who provide healthcare and social services in the community. Methods Items related to a CHSNA tool were developed and mapped with ICF codes. Experts validated the CHSNA system design and process using the Delphi method, and a pilot test of the initial version of the system was conducted. Results The following three steps of CHSNA were embedded in the system, which had a user-friendly screen and images: basic health assessment, life and activity assessment, and in-depth health assessment. The assessment results for the community residents were presented with visualized health profiles, including images, graphs, and an ICF model. Conclusions The developed CHSNA system can be used by healthcare professionals, social workers, and community residents to evaluate the reasoning underlying health and social needs, to facilitate the identification of more appropriate healthcare plans, and to guide community residents to receive the best healthcare services. A CHSNA system can improve the implementation of standardized terminology utilizing the ICF and the accuracy of needs assessments of community residents.

Purpose: The purpose of this study was to identify the relationship among the attitudes on exercise, physical activity and quality of life (QOL) in hemodialysis patients. Methods: A total of 42 patients in a hemodialysis unit participated in this study. Physical activity level was measured directly by 6 minute walking test and grip strength test. Structured questionnaires were also used for measuring their attitudes on exercise, physical activity and QOL. Participants` medical records were reviewed for obtaining their biochemical and clinical information. Statistical analysis was performed using Pearson correlation, and multiple liner regression. Results: A significant positive correlation between participants` attitudes and physical activity level measured by International Physical Activity Questionnaire (IPAQ) was found. And the physical activity level measured by Korea Activity Scale/Index (KASI) was significantly related to QOL. Conclusion: This study shows that QOL of the hemodialysis patients was significantly associated with their physical activity level.

Highlights? A pattern selection method called Expected Margin based Pattern Selection (EMPS) is proposed. ? EMPS reduces the training complexities of SVR for use as a response modeling dataset. ? The experimental results involving one real-world marketing dataset showed that EMPS improved SVR efficiency for response modeling. Two-stage response modeling, identifying respondents and then ranking them according to their expected profit, was proposed in order to increase the profit of direct marketing. For the second stage of two-stage response modeling, support vector regression (SVR) has been successfully employed due to its great generalization performances. However, the training complexities of SVR have made it difficult to apply to response modeling based on the large amount of data. In this paper, we propose a pattern selection method called Expected Margin based Pattern Selection (EMPS) to reduce the training complexities of SVR for use as a response modeling dataset with high dimensionality and high nonlinearity. EMPS estimates the expected margin for all training patterns and selects patterns which are likely to become support vectors. The experimental results involving 20 benchmark datasets and one real-world marketing dataset showed that EMPS improved SVR efficiency for response modeling.

Since semiconductor manufacturing consists of hundreds of processes, a faulty wafer detection system, which allows for earlier detection of faulty wafers, is required. statistical process control (SPC) and virtual metrology (VM) have been used to detect faulty wafers. However, there are some limitations in that SPC requires linear, unimodal and single variable data and VM underestimates the deviations of predictors. In this paper, seven different machine learning-based novelty detection methods were employed to detect faulty wafers. The models were trained with Fault Detection and Classification (FDC) data to detect wafers having faulty metrology values. The real world semiconductor manufacturing data collected from a semiconductor fab were tested. Since the real world data have more than 150 input variables, we employed three different dimensionality reduction methods. The experimental results showed a high True Positive Rate (TPR). These results are promising enough to warrant further study.

Samsung SDSThe purpose of virtual metrology (VM) in semiconductor manufacturing is to predict every wafer’s metrological values based on its process equipment data without an actual metrology. In this paper, we propose novelty detection-based reliability estimation models for VM in order to support flexible utilization of VM results. Because the proposed model can not only estimate the reliability of VM, but also identify suspicious process variables lowering the reliability, quality control actions can be taken selectively based on the reliance level and its causes. Based on the preliminary experimental results with actual semiconductor manufacturing process data, our models can successfully give a high reliance level to the wafers with small prediction errors and a low reliance level to the wafers with large prediction errors. In addition, our proposed model can give more detailed information by identifying the critical process variables and their relative impacts on the low reliability.

Response modeling has become a key factor to direct marketing. In general, there are two stages in response modeling. The first stage is to identify respondents from a customer database while the second stage is to estimate purchase amounts of the respondents. This paper focuses on the second stage where a regression, not a classification, problem is solved. Recently, several non-linear models based on machine learning such as support vector machines (SVM) have been applied to response modeling. However, there is a major difficulty. A typical training dataset for response modeling is so large that modeling takes very long, or, even worse, modeling may be impossible. Therefore, sampling methods have been usually employed in practice. However a sampled dataset usually leads to lower accuracy. In this paper, we employed an @e-tube based sampling for support vector regression (SVR) which leads to better accuracy than the random sampling method.

This paper presents a simple and accurate model for determining I on and I off of a double-gate FinFET with varying gate fin shapes. Simulations show that gate fin shape variation results in significant changes in the leakage and driving capability of the device. We perform TCAD simulations of double-gate FinFET structures in order to analyze the effect of the gate fin body thickness (T si ) variation on the electrical properties of the device. The thicknesses of the source and drain side are found to have different effects on the device. A simple model is proposed using the threshold voltage change due to the thickness variation along the gate fin. Simulation results show that the models match well with I on and I off within 1.3% and 4.8% errors, respectively. In addition, we propose an optimal fin body shape to reduce the leakage current while providing a similar driving current to that in the nominal FinFET.

Support Vector Machine (SVM) results in a good generalization performance by employing the Structural Risk Minimization (SRM) principle. However, one drawback is O(n3) training time complexity. In this paper, we propose a pattern selection method designed specifically for Support Vector Regression (SVR). In SVR training, only a few patterns called support vectors are used to construct the regression model while other patterns are not used at all. The proposed method tries to select patterns which are likely to become support vectors. With multiple bootstrap samples, we estimate the likelihood of each pattern to become a support vector. The proposed method automatically determines the appropriate number of patterns selected by estimating the expected number of support vectors. Through the experiments involving twenty datasets, the proposed method resulted in the best accuracy among the competing methods.

In induction motor control, power efficiency is an important factor to be considered. We attempt to achieve both high dynamic performance and maximum power efficiency by means of linear decoupling of rotor speed (or motor torque) and rotor flux. The induction motor with our controller possesses the input-output dynamic characteristics of a linear system such that the rotor speed (or motor torque) and the rotor flux are decoupled. The rotor speed responses are not affected by abrupt changes in the rotor flux and vice versa. The rotor flux need not be measured but is estimated by the well known flux simulator. The effect of large variation in the rotor resistance on the control performances is minimized by employing a parameter adaptation method. To illuminate the significance of our work, we present simulation and experimental results as well as mathematical performance analyses. In particular, our experimental work demonstrates that recently developed nonlinear feedback control theories are of practical use.

The performance of classification algorithms for differentiating among obstructive lung diseases based on features from texture analysis using HRCT (High Resolution Computerized Tomography) images was compared. HRCT can provide accurate information for the detection of various obstructive lung diseases, including centrilobular emphysema, panlobular emphysema and bronchiolitis obliterans. Features on HRCT images can be subtle, however, particularly in the early stages of disease, and image-based diagnosis is subject to inter-observer variation. To automate the diagnosis and improve the accuracy, we compared three types of automated classification systems, naA¯ve Bayesian classifier, ANN (Artificial Neural Net) and SVM (Support Vector Machine), based on their ability to differentiate among normal lung and three types of obstructive lung diseases. To assess the performance and cross-validation of these three classifiers, 5 folding methods with 5 randomly chosen groups were used. For a more robust result, each validation was repeated 100 times. SVM showed the best performance, with 86.5% overall sensitivity, significantly different from the other classifiers (one way ANOVA, p

An attempt is made to control induction motors with high power efficiency as well as high dynamic performance by utilizing recently developed nonlinear feedback control techniques. The controller consists of three subcontrollers: a saturation current controller, a decoupling controller, and a well-known flux simulator. The decoupling controller decouples rotor speed (or motor torque) and rotor flux linearly. The controller does not need the transformation between a d-q synchronously rotating frame and an x-y stator-fixed frame. It is computationally quite simple and does not depend on the rotor resistance, which varies widely with the machine temperature. Performance analysis and simulation results are presented. >