• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.73-95
• /
• 2021

This study uses Node2vec graph embedding method and Light GBM link prediction to explore undeveloped export candidate countries in Korea's food and beverage industry. Node2vec is the method that improves the limit of the structural equivalence representation of the network, which is known to be relatively weak compared to the existing link prediction method based on the number of common neighbors of the network. Therefore, the method is known to show excellent performance in both community detection and structural equivalence of the network. The vector value obtained by embedding the network in this way operates under the condition of a constant length from an arbitrarily designated starting point node. Therefore, it has the advantage that it is easy to apply the sequence of nodes as an input value to the model for downstream tasks such as Logistic Regression, Support Vector Machine, and Random Forest. Based on these features of the Node2vec graph embedding method, this study applied the above method to the international trade information of the Korean food and beverage industry. Through this, we intend to contribute to creating the effect of extensive margin diversification in Korea in the global value chain relationship of the industry. The optimal predictive model derived from the results of this study recorded a precision of 0.95 and a recall of 0.79, and an F1 score of 0.86, showing excellent performance. This performance was shown to be superior to that of the binary classifier based on Logistic Regression set as the baseline model. In the baseline model, a precision of 0.95 and a recall of 0.73 were recorded, and an F1 score of 0.83 was recorded. In addition, the light GBM-based optimal prediction model derived from this study showed superior performance than the link prediction model of previous studies, which is set as a benchmarking model in this study. The predictive model of the previous study recorded only a recall rate of 0.75, but the proposed model of this study showed better performance which recall rate is 0.79. The difference in the performance of the prediction results between benchmarking model and this study model is due to the model learning strategy. In this study, groups were classified by the trade value scale, and prediction models were trained differently for these groups. Specific methods are (1) a method of randomly masking and learning a model for all trades without setting specific conditions for trade value, (2) arbitrarily masking a part of the trades with an average trade value or higher and using the model method, and (3) a method of arbitrarily masking some of the trades with the top 25% or higher trade value and learning the model. As a result of the experiment, it was confirmed that the performance of the model trained by randomly masking some of the trades with the above-average trade value in this method was the best and appeared stably. It was found that most of the results of potential export candidates for Korea derived through the above model appeared appropriate through additional investigation. Combining the above, this study could suggest the practical utility of the link prediction method applying Node2vec and Light GBM. In addition, useful implications could be derived for weight update strategies that can perform better link prediction while training the model. On the other hand, this study also has policy utility because it is applied to trade transactions that have not been performed much in the research related to link prediction based on graph embedding. The results of this study support a rapid response to changes in the global value chain such as the recent US-China trade conflict or Japan's export regulations, and I think that it has sufficient usefulness as a tool for policy decision-making.

• Korean Journal of Psychosomatic Medicine
• /
• v.19 no.2
• /
• pp.66-73
• /
• 2011

Objectives:The present study investigates the status of inpatient psychiatric consultations at a general hospital in order to find factors that contribute to treatment compliance related to psychiatric consultations. Methods:The subjects were 333 patients who were hospitalized at Korea University Medical Center Ansan Hospital from 1 September 2009 to 31 July 2010.The patients were referred for psychiatric consultation during hospitalization. This study investigates demographic data, request department, referral causes, requestor, psychiatric history and diagnosis, andpsychiatric treatment compliance. Treatment compliance was defined as whether or not the patient had accepted psychiatric treatment during hospitalization or outpatient department(OPD) follow-up. This study ascertains the factors that have impact on compliance, by taking binary logistic regression with compliance and other variables. Results:Among the patients that were offered psychiatric treatment during hospitalization(N=310), treatment compliance was 82.9%. Among the patients that were offered OPD treatment(N=111), compliance was 55.8%. Elderly group(>65 years) showed better compliance to treatment during hospitalization than the younger patient group(OR=4.838, p=0.004). Patients with secondary psychiatric disorders showed better OPD follow-up compliance than patients with secondary psychiatric disorders(OR=8.520, p=.008). Conclusion:Elderly patients showed better compliance for psychiatric treatment during hospitalization. However they commonly have disorders such as delirium and mood disorders that have impact on the patient's physical state, hence further active measures should be carried out. Patients referred due to primary psychiatric disorders showed poor OPD compliance. Therefore clinicians have to suggest multidisciplinary interventions that will improve treatment compliance of such patients.

• Clinical and Experimental Pediatrics
• /
• v.52 no.9
• /
• pp.1021-1028
• /
• 2009

Purpose:Medical therapy is the initial treatment for children with Graves disease to avoid complications of other treatments. However, optimal treatment for childhood Graves disease is controversial because most patients require relatively long periods of medical therapy and relapse is common after medication discontinuation. Therefore, this study aimed to search clinical or biochemical characteristics that could be used as remission predictors in Graves disease. Methods:We retrospectively studied children diagnosed with Graves disease, treated with anti-thyroid agents, and observed for at least 3 years. Patients were categorized into remission and non-remission groups, and the groups were compared to determine the variables that were predictive of achieving remission. Results:Sixty-four patients were enrolled, of which 37 (57.8%) achieved remission and 27 (42.2%) could not achieve remission until the last visit. Normalization of thyroid-stimulating hormone-binding inhibitory immunoglobulin (TBII) after treatment was faster in the remission group than in the non-remission group (remission group, $15.5{\pm}12.07$ vs. non-remission group, $41.69{\pm}35.70$ months). Thyrotropin-releasing hormone (TRH) stimulation tests were performed in 28 patients. Only 2 (8.3%) of 26 patients who showed normal or hyper-response in TRH stimulation test relapsed. Binary logistic regression analysis identified rapid achievement of TBII normalization after treatment as a significant predictor of remission. Six percent of patients achieved remission within 3 years and 55.8% achieved it within 6 years. Conclusion:Rapid achievement of TBII normalization can be a predictor of remission in childhood Graves disease. The TRH stimulation test can be a predictor of maintenance of remission.

• Asia pacific journal of information systems
• /
• v.19 no.2
• /
• pp.157-178
• /
• 2009

Corporate credit rating is a very important factor in the market for corporate debt. Information concerning corporate operations is often disseminated to market participants through the changes in credit ratings that are published by professional rating agencies, such as Standard and Poor's (S&P) and Moody's Investor Service. Since these agencies generally require a large fee for the service, and the periodically provided ratings sometimes do not reflect the default risk of the company at the time, it may be advantageous for bond-market participants to be able to classify credit ratings before the agencies actually publish them. As a result, it is very important for companies (especially, financial companies) to develop a proper model of credit rating. From a technical perspective, the credit rating constitutes a typical, multiclass, classification problem because rating agencies generally have ten or more categories of ratings. For example, S&P's ratings range from AAA for the highest-quality bonds to D for the lowest-quality bonds. The professional rating agencies emphasize the importance of analysts' subjective judgments in the determination of credit ratings. However, in practice, a mathematical model that uses the financial variables of companies plays an important role in determining credit ratings, since it is convenient to apply and cost efficient. These financial variables include the ratios that represent a company's leverage status, liquidity status, and profitability status. Several statistical and artificial intelligence (AI) techniques have been applied as tools for predicting credit ratings. Among them, artificial neural networks are most prevalent in the area of finance because of their broad applicability to many business problems and their preeminent ability to adapt. However, artificial neural networks also have many defects, including the difficulty in determining the values of the control parameters and the number of processing elements in the layer as well as the risk of over-fitting. Of late, because of their robustness and high accuracy, support vector machines (SVMs) have become popular as a solution for problems with generating accurate prediction. An SVM's solution may be globally optimal because SVMs seek to minimize structural risk. On the other hand, artificial neural network models may tend to find locally optimal solutions because they seek to minimize empirical risk. In addition, no parameters need to be tuned in SVMs, barring the upper bound for non-separable cases in linear SVMs. Since SVMs were originally devised for binary classification, however they are not intrinsically geared for multiclass classifications as in credit ratings. Thus, researchers have tried to extend the original SVM to multiclass classification. Hitherto, a variety of techniques to extend standard SVMs to multiclass SVMs (MSVMs) has been proposed in the literature Only a few types of MSVM are, however, tested using prior studies that apply MSVMs to credit ratings studies. In this study, we examined six different techniques of MSVMs: (1) One-Against-One, (2) One-Against-AIL (3) DAGSVM, (4) ECOC, (5) Method of Weston and Watkins, and (6) Method of Crammer and Singer. In addition, we examined the prediction accuracy of some modified version of conventional MSVM techniques. To find the most appropriate technique of MSVMs for corporate bond rating, we applied all the techniques of MSVMs to a real-world case of credit rating in Korea. The best application is in corporate bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. For our study the research data were collected from National Information and Credit Evaluation, Inc., a major bond-rating company in Korea. The data set is comprised of the bond-ratings for the year 2002 and various financial variables for 1,295 companies from the manufacturing industry in Korea. We compared the results of these techniques with one another, and with those of traditional methods for credit ratings, such as multiple discriminant analysis (MDA), multinomial logistic regression (MLOGIT), and artificial neural networks (ANNs). As a result, we found that DAGSVM with an ordered list was the best approach for the prediction of bond rating. In addition, we found that the modified version of ECOC approach can yield higher prediction accuracy for the cases showing clear patterns.

• Tuberculosis and Respiratory Diseases
• /
• v.64 no.6
• /
• pp.414-421
• /
• 2008

Background: Recently, in addition to multi-drug resistant tuberculosis (MDR-TB), extensively drug-resistant tuberculosis (XDR-TB) has become rapidly growing public health threat. This study examined the clinical differences between pulmonary TB patients with extensively drug resistance (XDR) and multi-drug resistance (MDR) at the National Medical Center in Korea in order to determine the clinical characteristics associated more with XDR-TB than MDR-TB. Methods: Patients who received a diagnosis of culture-confirmed pulmonary TB and a drug sensitivity test (DST) for anti-TB drugs at the National Medical Center between January 2000 and August 2007 were enrolled in this study. The patients were identified into the XDR-TB or MDR-TB group according to the DST results. The clinical characteristics were reviewed retrospectively from the medical records. Statistical analysis for the comparisons was performed using a ${\chi}^2$-test, independent samples t-test or binary logistic regression where appropriate. Results: A total 314 patients with culture-confirmed pulmonary TB were included. Among them, 18 patients (5.7%) had XDR-TB and 69 patients (22%) had MDR-TB excluding XDR-TB. A comparison of the clinical characteristics, revealed the XDR-TB group to have a higher frequency of a prior pulmonary resection for the treatment of TB (odds ratio [OR], 3.974; 95% confidence interval [CI], 1.052~15.011; P value 0.032) and longer average previous treatment duration with anti-TB drugs, including a treatment interruption period prior to the diagnosis of XDR, than the MDR-TB group (XDR-TB group, 72.67 months; MDR-TB group, 13.09 months; average treatment duration difference between two groups, 59.582 months; 95% CI, 31.743~87.420; P value, 0.000). In addition, a longer previous treatment duration with anti-TB drugs was significantly associated with XDR-TB (OR, 1.076; 95% CI, 1.038~1.117; P value, 0.000). A comparison of the other clinical characteristics revealed the XDR-TB group to have a higher frequency of male gender, diabetes mellitus (DM), age under 45, treatment interruption history, cavitations on simple chest radiograph and positive result of sputum AFB staining at the time of diagnosis of XDR. However, the association was not statistically significant. Conclusion: Pulmonary TB patients with XDR have a higher frequency of a prior pulmonary resection and longer previous treatment duration with anti-TB drugs than those with MDR. In addition, a longer previous treatment duration with anti-TB drugs is significantly associated with XDR-TB.

• Journal of dental hygiene science
• /
• v.14 no.3
• /
• pp.311-318
• /
• 2014

The purpose of this study was to look into the perception, experience, treatment of early childhood caries (ECC) and influential factors of perception in order to provide basic data useful for preventing the ECC by examining the relationship between oral health of young children in infancy and mother. In this study, 277 mothers were surveyed who had children in children under 6 years old and visiting the pediatrics, day care center, and pediatric dental clinics located in Daegu and Gyeongsangbuk-do from July 10, 2013 to September 5 of the same year. The results obtained from the survey were analyzed through chi-square test, t-test, and binary logistic regression analysis by using the SPSS 18.0, a statistical program. The results of analysis showed that ECC in children under 6 years old was associated with mother's age, education background of mothers, number of children and monthly income and had a significant correlation with mother's oral health-related knowledge and oral health care of their children. Thus, it would be necessary to develop oral health education programs and implement such oral health education programs at a national level on a regular basis for the mothers of young children in infancy and would-be mothers in order to reduce the ECC in infancy and promote oral health.

• The Korean Society of Law and Medicine
• /
• v.19 no.2
• /
• pp.235-261
• /
• 2018

The Medical law stipulates regulations about the physician's duty to inform to contribute to patient's self-determination. This law was most recently revised on December 20, 2016, and came into effect on June 21, 2017. There has been much controversy about this, and it has been questioned whether or not it will be effective for physicians to comply with the duty to inform. Therefore, this study investigated perceptions of physicians of whether they observed the duty to inform and their legal judgment about that duty, and analyzed how the revision of the medical law may have affected the legal cognition of physician's duty to inform. This study was conducted through an online questionnaire survey involving 109 physicians over 2 weeks from March 29 to April 12, 2018, and 108 of the collected data were used for analysis. The questionnaire was developed by revising and supplementing the previous research (Lee, 2004). It consisted of 41 items, including 26 items related to the experience of and legal judgment about the duty to inform, 6 items related to awareness of revised medical law, and 9 items on general characteristics. The data were analyzed using SAS 9.4 program and descriptive statistics, Chi-square test, Fisher's exact test and Binary logistic regression were performed. The results are as follows. • Out of eight situations, the median number of situations that did not fulfill the duty to inform was 5 (IQR, 4-6). In addition, 12 respondents (11%) answered that they did not fulfill the duty to inform in all eight cases, while only one (1%) responded that he/she performed explanation obligations in all cases. • The median number of the legal judgment score on the duty to inform was 8 out of 13 (IQR, 7-9), and the scores ranged from a minimum of 4 (4 respondents) to a maximum of 11 (3 respondents). • More than half of the respondents (n=26, 52%) were unaware of the revision of the medical law, 27 (25%) were aware of the fact that the medical law had been revised, 20(18%) had a rough knowledge of the contents of the law, and only 5(5%) said they knew the contents of the law in detail. The level of awareness of the revised medical law was statistically significant difference according to respondents' sex (p<.49), age (p<.0001), career (p<.0001), working type (p<.024), and department (p<.049). • There was no statistically significant relationship between the level of awareness of the revised medical law and the level of legal judgment on the duty to inform. These results suggest that efforts to improve the implementation and cognition of physician's duty to inform are needed, and it is difficult to expect a direct positive effect from the legal regulations per se. Considering the distinct characteristics of medical institutions and hierarchical organizational culture of physicians, it is necessary to develop a credible guideline on the duty to inform within the medical system, and to strengthen the education of physicians about their duty to inform and its purpose.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.4
• /
• pp.135-151
• /
• 2021

This research attempts to analyze the factors influencing the entrepreneurial intention of employees in manufacturing field. In particular, key factors of entrepreneurship and personal characteristics explain a significant association with the intention to start-up. And study whether R&D support from public enterprise adjusts intention to entrepreneurial Intention. This study conducted a online survey on 292 small and medium-sized enterprise manufacturing employees in May 2020. Using linear regression model and binary logistic model. The main study results are the following: First, among the key factors(innovativeness, proactiveness, risk-taking) of entrepreneurship, proactiveness hardly influenced the opportunity competency. Second, among the factors(risk-taking propensity, locus of control, tolerance for ambiguity) of personal characteristics, locus of control hardly influenced the opportunity competency. Third, opportunity competency(opportunity recognition and opportunity evaluation) had positive influence to entrepreneurial intention. Fourth, the study investigated the mediated effect of opportunity competency. The result showed that among the factors of entrepreneurship and personal characteristics, only two factors that are proactiveness and locus of control were not mediated by opportunity competency. and opportunity evaluation was acted as a mediator between proactiveness and entrepreneurial Intention, compared with opportunity recognition. Lastly, public enterprise's R&D supporting moderated the entrepreneurial intention). Based on the result, the study showed that first, the key factors of entrepreneurship except for proactiveness and personal characteristics(risk-taking propensity, locus of control, tolerance for ambiguity) except for locus of control affect the intention to start-up, repeatedly. This results are explained that employees have not started a business yet. Second, research on start-up suggests the need to analyze factors differentiated before and after the start-ups. Based on the results, entrepreneurship and personal characteristics show that study on the effects of start-up intentions should be carried out before and after the actual start-up takes place, and can be used as effective data in policies to promoting start-ups in manufacturing field.