• Journal of Navigation and Port Research
• /
• v.48 no.3
• /
• pp.200-205
• /
• 2024

This paper examines the impact of ESG activities of domestic shipping companies on the cost of debt. It is known that companies with large information asymmetry tend to have high costs of debt. Corporate ESG activities have been identified as an effective means of reducing information asymmetry. By actively engaging in ESG activities, companies can lower the cost of debt by reducing information asymmetry. Therefore, this study aims to investigate whether these mechanisms, which have been observed in previous studies, also apply to domestic shipping companies. Multiple regression analysis is conducted on KOSP I-listed shipping companies from2010 to 2022. The cost of debt is set as the dependent variable, while the ESG rating is used as the explanatory variable. The analysis reveals that companies with a high level of ESG activities generally have a lower cost of debt. However, it is important to note that ESG activities of shipping companies do not seem to have a significant impact on their cost of debt. In fact, the level of ESG activities among domestic shipping companies is not particularly high (Hong, 2024). Despite these findings, domestic shipping companies should still strive for sustainable management to adapt to the rapidly changing business environment and meet the demands of the modern era. ESG management is a representative method for achieving sustainability. Therefore, shipping companies should not only focus on reducing the cost of debt but also on opening up the closed industry culture and communicating with capital market participants for sustainable growth. It is crucial for these companies to listen to the voices of stakeholders and embrace a holistic approach to sustainability.

• Journal of dental hygiene science
• /
• v.5 no.1
• /
• pp.39-43
• /
• 2005

This study was carried out to provide fundamental data for an examination of a health promotion program by determining the influence of the Health Promoting Lifestyle Profile on the quality of life of female dental hygienists. The sample was selected from the population of 1,148 who were registered in the Association of Seoul Dental Hygienists. 800 subjects were randomly selected from 25 districts in Seoul. The data was collected by calling the dental hygienists to, explaining the contents and objective of our study, and sending them a questionnaire by post. The questionnaire consists of the total number of 97 questions: 62 questions on the Health Promoting Lifestyle Profile, 26 questions on the quality of life and 9 general characteristics questions. The data was collected from August 16 to October 15, 2004. Out of 800 subjects, 481(60.1%) completed the questionnaires. For statistical analysis, the frequency, percentage, arithmetic mean, ANOVA, and multiple regression analysis, were analyzed using the SAS 8.1 Analysis program. The significance level was set to 0.05. The results of this study were as follows: First, The average score of the subjects' quality of life was 3.1. For the sub-categories, it was shown that the degree of satisfaction on the condition of society was the highest at 3.2, and the degree of satisfaction on the condition of the individuals was the lowest at 3.1. The average score of the Health Promoting Lifestyle Profile variable was 2.5. For the sub-categories, it was shown that the degree of sanitary life was at 3.2, and degree of the professional health maintenance was the lowest at 1.7. Second, There were significant differences in the Quality of Life benefits of action with the general characteristics. There were significant differences in age, educational level, income, marital status, career, and Perceived Health Status. There were significant difference in Health Promoting Lifestyle Profile benefits of action with the general characteristics. There were significant differences in terms of age, educational level, income, marital status, career, and the Perceived Health Status. Finally, The stepwise multiple regression analysis revealed that the powerful predictors were Health Promoting Lifestyle Profile, income and the Perceived Health Status. These factors accounted for 37.6% of the variance in the Quality of Life patterns. As the subjects were limited to dental hygienists in Seoul, care should be taken when applying these results to all dental hygienists in Korea. In order to generalize the study, a large number of subjects selected from all regions in Korea will be needed.

• Tuberculosis and Respiratory Diseases
• /
• v.56 no.6
• /
• pp.600-610
• /
• 2004

Background : The plasma B-type natriuretic peptide(BNP) concentration increases with the degree of pulmonary hypertension in patients with chronic respiratory disease. The aim of this study was to examine the prognostic role of BNP in the acute exacerbation of chronic obstructive lung disease (COPD). Method : We selected 67 patients who were admitted our hospital because of an acute exacerbation of COPD. Their BNP levels were checked on admission at the Emergency Department. Their medical records were analyzed retrospectively. The patients were divided into two groups according to their in-hospital mortality. The patients' medical history, comobidity, exacerbation type, blood gas analysis, pulmonary function, APACHE II severity score and plasma BNP level were compared. Results : Multiple logistic regression analysis identified three independent predictors of mortality: $FEV_1$, APACHE II score and plasma BNP level. The decedents group showed a lower $FEV_1$($28{\pm}7$ vs. $37{\pm}15%$, p=0.005), a higher APACHE II score($22.4{\pm}6.1$ vs. $15.8{\pm}4.7$, p=0.000) and a higher BNP level ($201{\pm}116$ vs. $77{\pm}80pg/mL$, p=0.000) than the sSurvivors group. When the BNP cut-off level was set to 88pg/mL using the receiver operating characteristic curve, the sensitivity was 90% and the specificity was 75% in differentiating between the survivors and decedents. On Fisher's exact test, the odds ratio for mortality was 21.2 (95% CI 2.49 to 180.4) in the patients with a BNP level > 88pg/mL. Conclusion : The plasma BNP level might be a predictor of mortality in an acute exacerbation of COPD as well as the $FEV_1$ and APACHE II score.

• Journal of the Society of Disaster Information
• /
• v.11 no.2
• /
• pp.191-202
• /
• 2015

The purpose of this study was to examine the influence of the disaster perception and satisfaction level of volunteer fire-fighters on the continuity of their volunteering. The 163 subjects in this study were selected from male and female volunteer fire-fighters who resided in urban and rural regions in Gangwon Province. After a this survey was conducted, the collected data were analyzed by a statistical package SPSS WIN 20.0, and frequency analysis, correlation analysis and multiple regression analysis were made. The level of statistical significance was all set at p<.05. The findings of the study were as follows: First, as for the correlation of disaster perception, satisfaction and volunteering continuity, perception of disaster countermeasures, satisfaction and the level of participation were negatively correlated with one another, and there was a positive correlation among disaster training, disaster preparation, regional disaster, the period of volunteering, and will of persistent volunteering. Second, as for perception of disaster, the volunteer fire-fighters were asked a question about disaster countermeasures, and the largest group replied they were partially aware of the countermeasures. Concerning questions about disaster training/education experience and triage, the biggest group replied they underwent the training and knew about triage on the whole. Regarding questions on the emergency contact system and emergency work schedule, they knew about the two in general. As to a question on the occurrence of human disaster, the greatest group answered that they knew about it yet not well. Third, in regard to the impact of satisfaction level on volunteering continuity, the period of volunteering was affected by needs for experience, social contact and social recognition among the subfactors of satisfaction level, and will of persistent volunteering was under the influence of social contact and achievement needs. The level of persistent volunteering was affected only by needs for experience and achievement needs.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.27 no.1
• /
• pp.1-12
• /
• 1991

An experiment has been carefully designed and performed to verify the theory for the echointergration technique of estimating the density of fish school by the use of steel spheres in a laboratory tank. The spheres used to simulate a fish school were randomly distributed throughout the insonified volume to produce the acoustic echoes similar to those scattered from real fish schools. The backscattered echoes were measured as a function of target density at tow frequencies of 50kHz and 200kHz. Data acquisition, processing and analysis were performed by means of the microcomputer-based sonar-echo processor including a FFT analyzer. Acoustic scattering characteristics of a 36cm mackerel was investigated by measuring fish echoes with frequencies ranging from 47.8kHz to 52.0kHz. The fluctuation of bottom echoes caused by the effects of fish-school attenuation and multiple scattering which occurred in dense aggregations of fishes was also examined by analyzing the echograms of sardine schools obtained by a 50kHz telesounder in the set-net's bagnet, and the echograms obtained by a scientific echo sounder of 50kHz in the East China Sea, respectively. The results obtained can be summarized as follows: 1. The measured and the calculated echo shapes on the steel sphere used to simulate a fish school were in close agreement. 2. The waveform and amplitude of echo signals by a mackerel without swimbladder fluctuated irregularly with the measuring frequency. 3. When a collection of 30 targets/m super(3) lied the shadow region behind another collection of 5 targets/m super(3), the mean losses in echo energy for the 30 targets/m super(3) were about -0.4dB at 50kHz and about -0.2dB at 200kHz, respectively. 4. In the echograms obtained in the East China Sea, the bottom echoes fluctuated remarkably when the dense aggregations of fish appeared between transducer and seabed. Especially, in the case of the echograms of sardine school obtained in a set-net's bagnet, the disappearance of bottom echoes and the lengthening of the echo trace by fish aggregations were observed. Then the mean density of the sardine school was estimated as 36 fish/m super(3). It suggests that when the distribution density of fishes in oceans is greater than this density, the effects of fish-school attenuation and multiple scattering must be taken into account as a possible source of error in fish abundance estimates. 5. The relationship between mean backscattering strength (, dB) and target density ($\rho$, No./m super(3)) were expressed by the equations: =-46.2+13.7 Log($\rho$) at 50kHz and =-43.9+13.4 Log($\rho$) at 200kHz. 6. The difference between the experimentally derived number and the actual number of targets gradually decreased with an increase in the target density and was within 20% when the density was 30 targets/m super(3). From these results, we concluded that when the number of targets in the insonified volume is large, the validity of the echo-integration technique of estimating the density of fish schools could be expected.

• (The) Research of the performance art and culture
• /
• no.41
• /
• pp.5-38
• /
• 2020

The current state of performance halls in South Korea is closely related to the performance art and culture of the nation as the culture of putting on and enjoying a performance is deeply rooted in public culture and arts halls representing each area at the local government level. Today, public culture and arts halls have multiple management purposes, and the subjects of their management are in the public domain including the central and local governments or investment and donation foundations in overwhelming cases. Public culture and arts halls thus have close correlations with the institutional aspect of cultural policies as the objects of culture and art policies at the central and local government level. The full-blown era of public culture and arts halls opened up in the 1980s~1990s, during which multi-purpose performance halls of a similar structure became universal around the nation. Public culture and arts halls of the uniform shape were distributed around the nation with no premise of genre characteristics or local environments for arts, and this was attributed to the cultural policies of the military regime. The Park Chung-hee regime proclaimed Yusin that was beyond the Constitution and enacted the Culture and Arts Promotion Act(September, 1972), which was the first culture and arts act in the nation. Based on the act, a five-year plan for the promotion of culture and arts(1973) was made and led to the construction of cultural facilities. "Public culture and arts" halls or "culture" halls were built to serve multiple purposes around the nation because the Culture and Arts Promotion Act, which is called the starting point of the nation's legal system for culture and arts, defined "culture and arts" as "matters regarding literature, art, music, entertainment, and publications." The definition became a ground for the current "multi-purpose" concept. The organization of Ministry of Culture and Public Information set up a culture and administration system to state its supervision of "culture and arts" and distinguish popular culture from the promotion of arts. During the period, former President Park exhibited his perception of "culture=arts=culture and arts" in his speeches. Arts belonged to the category of culture, but it was considered as "culture and arts." There was no department devoted to arts policies when the act was enacted with a broad scope of culture accepted. This ambiguity worked as a mechanism to mobilize arts in ideological utilizations as a policy. Against this backdrop, the Sejong Center for the Performing Arts, a multi-purpose performance hall, was established in 1978 based on the Culture and Arts Promotion Act under the supervision of Ministry of Culture and Public Information. There were, however, conflicts of value over the issue of accepting the popular music among the "culture and arts = multiple purposes" of the system, "culture ≠ arts" of the cultural organization that pushed forward its establishment, and "culture and arts = arts" perceived by the powerful class. The new military regime seized power after Coup d'état of December 12, 1979 and failed at its culture policy of bringing the resistance force within the system. It tried to differentiate itself from the Park regime by converting the perception into "expansion of opportunities for the people to enjoy culture" to gain people's supports both from the side of resistance and that of support. For the Chun Doo-hwan regime, differentiating itself from the previous regime was to secure legitimacy. Expansion of opportunities to enjoy culture was pushed forward at the level of national distribution. This approach thus failed to settle down as a long-term policy of arts development, and the military regime tried to secure its legitimacy through the symbolism of hardware. During the period, the institutional ground for public culture and arts halls was based on the definition of "culture and arts" in the Culture and Arts Promotion Act enacted under the Yusin system of the Park regime. The "multi-purpose" concept, which was the management goal of public performance halls, was born based on this. In this context of the times, proscenium performance halls of a similar structure and public culture and arts halls with a similar management goal were established around the nation, leading to today's performance art and culture in the nation.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.99-112
• /
• 2010

Ensemble learning is a method for improving the performance of classification and prediction algorithms. It is a method for finding a highly accurateclassifier on the training set by constructing and combining an ensemble of weak classifiers, each of which needs only to be moderately accurate on the training set. Ensemble learning has received considerable attention from machine learning and artificial intelligence fields because of its remarkable performance improvement and flexible integration with the traditional learning algorithms such as decision tree (DT), neural networks (NN), and SVM, etc. In those researches, all of DT ensemble studies have demonstrated impressive improvements in the generalization behavior of DT, while NN and SVM ensemble studies have not shown remarkable performance as shown in DT ensembles. Recently, several works have reported that the performance of ensemble can be degraded where multiple classifiers of an ensemble are highly correlated with, and thereby result in multicollinearity problem, which leads to performance degradation of the ensemble. They have also proposed the differentiated learning strategies to cope with performance degradation problem. Hansen and Salamon (1990) insisted that it is necessary and sufficient for the performance enhancement of an ensemble that the ensemble should contain diverse classifiers. Breiman (1996) explored that ensemble learning can increase the performance of unstable learning algorithms, but does not show remarkable performance improvement on stable learning algorithms. Unstable learning algorithms such as decision tree learners are sensitive to the change of the training data, and thus small changes in the training data can yield large changes in the generated classifiers. Therefore, ensemble with unstable learning algorithms can guarantee some diversity among the classifiers. To the contrary, stable learning algorithms such as NN and SVM generate similar classifiers in spite of small changes of the training data, and thus the correlation among the resulting classifiers is very high. This high correlation results in multicollinearity problem, which leads to performance degradation of the ensemble. Kim,s work (2009) showedthe performance comparison in bankruptcy prediction on Korea firms using tradition prediction algorithms such as NN, DT, and SVM. It reports that stable learning algorithms such as NN and SVM have higher predictability than the unstable DT. Meanwhile, with respect to their ensemble learning, DT ensemble shows the more improved performance than NN and SVM ensemble. Further analysis with variance inflation factor (VIF) analysis empirically proves that performance degradation of ensemble is due to multicollinearity problem. It also proposes that optimization of ensemble is needed to cope with such a problem. This paper proposes a hybrid system for coverage optimization of NN ensemble (CO-NN) in order to improve the performance of NN ensemble. Coverage optimization is a technique of choosing a sub-ensemble from an original ensemble to guarantee the diversity of classifiers in coverage optimization process. CO-NN uses GA which has been widely used for various optimization problems to deal with the coverage optimization problem. The GA chromosomes for the coverage optimization are encoded into binary strings, each bit of which indicates individual classifier. The fitness function is defined as maximization of error reduction and a constraint of variance inflation factor (VIF), which is one of the generally used methods to measure multicollinearity, is added to insure the diversity of classifiers by removing high correlation among the classifiers. We use Microsoft Excel and the GAs software package called Evolver. Experiments on company failure prediction have shown that CO-NN is effectively applied in the stable performance enhancement of NNensembles through the choice of classifiers by considering the correlations of the ensemble. The classifiers which have the potential multicollinearity problem are removed by the coverage optimization process of CO-NN and thereby CO-NN has shown higher performance than a single NN classifier and NN ensemble at 1% significance level, and DT ensemble at 5% significance level. However, there remain further research issues. First, decision optimization process to find optimal combination function should be considered in further research. Secondly, various learning strategies to deal with data noise should be introduced in more advanced further researches in the future.

• Asia pacific journal of information systems
• /
• v.23 no.4
• /
• pp.103-128
• /
• 2013

When a consumer needs a product or service and multiple sellers are available online, the process of selecting a seller to buy online from is complex since the process involves many behavioral dimensions that have to be taken into account. As a part of this selection process, consumers may set minimum trust expectation that can be used to screen out less trustworthy sellers. In the previous research, the level of consumers' trust expectation has been anchored on two important factors: product involvement and perceived risk. Product involvement refers to the extent to which a consumer perceives a specific product important. Thus, the higher product involvement may result in the higher trust expectation in sellers. On the other hand, other related studies found that when consumers perceived a higher level of risk (e.g., credit card fraud risk), they set higher trust expectation as well. While abundant research exists addressing the relationship between product involvement and perceived risk, little attention has been paid to the integrative view of the link between the two constructs and their impacts on the trust expectation. The present paper is a step toward filling this research gap. The purpose of this paper is to understand the process by which a consumer chooses an online merchant by examining the relationships among product involvement, perceived risk, trust expectation, and intention to buy from an e-tailer. We specifically focus on the mediating role of perceived risk in the relationships between enduring product involvement and the trust expectation. That is, we question whether product involvement affects the trust expectation directly without mediation or indirectly mediated by perceived risk. The research model with four hypotheses was initially tested using data gathered from 635 respondents through an online survey method. The structural equation modeling technique with partial least square was used to validate the instrument and the proposed model. The results showed that three out of the four hypotheses formulated were supported. First, we found that the intention to buy from a digital storefront is positively and significantly influenced by the trust expectation, providing support for H4 (trust expectation ${\rightarrow}$ purchase intention). Second, perceived risk was found to be a strong predictor of trust expectation, supporting H2 as well (perceived risk ${\rightarrow}$ trust expectation). Third, we did not find any evidence of direct influence of product involvement, which caused H3 to be rejected (product involvement ${\rightarrow}$ trust expectation). Finally, we found significant positive relationship between product involvement and perceived risk (H1: product involvement ${\rightarrow}$ perceived risk), which suggests that the possibility of complete mediation of perceived risk in the relationship between enduring product involvement and the trust expectation. As a result, we conducted an additional test for the mediation effect by comparing the original model with the revised model without the mediator variable of perceived risk. Indeed, we found that there exists a strong influence of product involvement on the trust expectation (by intentionally eliminating the variable of perceived risk) that was suppressed (i.e., mediated) by the perceived risk in the original model. The Sobel test statistically confirmed the complete mediation effect. Results of this study offer the following key findings. First, enduring product involvement is positively related to perceived risk, implying that the higher a consumer is enduringly involved with a given product, the greater risk he or she is likely to perceive with regards to the online purchase of the product. Second, perceived risk is positively related to trust expectation. A consumer with great risk perceptions concerning the online purchase is likely to buy from a highly trustworthy online merchant, thereby mitigating potential risks. Finally, product involvement was found to have no direct influence on trust expectation, but the relationship between the two constructs was indirect and mediated by the perceived risk. This is perhaps an important theoretical integration of two separate streams of literature on product involvement and perceived risk. The present research also provides useful implications for practitioners as well as academicians. First, one implication for practicing managers in online retail stores is that they should invest in reducing the perceived risk of consumers in order to lower down the trust expectation and thus increasing the consumer's intention to purchase products or services. Second, an academic implication is that perceived risk mediates the relationship between enduring product involvement and trust expectation. Further research is needed to elaborate the theoretical relationships among the constructs under consideration.

• Restorative Dentistry and Endodontics
• /
• v.24 no.4
• /
• pp.560-569
• /
• 1999

Blood supply rather than nerve supply implies pulp vitality. To evaluate pulp vitality clinically, electric pulp test and thermal test which are based on sensory nerve response have been used in addition to many auxiliary data such as past dental history, visual inspection, radiographic examination, percussion, palpation and transillumination test. However, reactivity of the nerves to the stimulation is not synonymous with normalcy. Therefore measurement of pulpal blood flow using a laser Doppler flowmeter became a new trial to test the pulp vitality. The purpose of the present study was to evaluate normal pulpal blood flow level of maxillary teeth in adult to provide a guideline in determining the vitality of dental pulp. Pulpal blood flow was measured in maxillary central and lateral incisors, canines, first and second premolars and first molars of seventy nine adults of 22 - 30 years old using a laser Doppler flowmeter (PeriFlux 4001, Perimed Co., Stockholm, Sweden, 780 nm infrared laser, 1mW). For directly-made splints, silicone rubber impressions were taken directly from the mouth. For indirectly-made splints, alginate impressions were taken from the mouth and stone cast were made. After making depressions on the buccal surfaces of the cast teeth to indicate the hole positions, second impressions with vinyl polysyloxane putty were taken from the cast. Holes for the laser probes were made at the putty impressions 4mm above the gingival level. Laser probe (PF416 dental probe, 1.5mm) was inserted in the prepared hole and the splint was set in the mouth. After 10 minutes of patient relaxing, pulpal blood flow was recorded for 5 minutes on each tooth. The recorded flow was saved in the computer and calculated with a software 'Perisoft' version 5.1. Pulpal blood flow was also recorded in six teeth of five individuals with no response to electric pulp test and cold test, with periapical radiolucency, or with history of root canal treatment to compare with nonvital teeth. The difference between the mean flow values of each group of teeth were analyzed using one-way ANOVA and Duncan's Multiple Range test. The results were as follows: 1. The average pulpal blood flow values of all the tested teeth of each location were between 9 - 16 Perfusion Unit. Pulpal blood flow value was highest in maxillary lateral incisors, followed by first premolars, second premolars, canines, central incisors, and then first molars (p<0.01). 2. In six anterior teeth, indirectly-made splint group showed higher pulpal blood flow values than directly-made splint group (p<0.01). In posterior teeth, however, there was no significant flow value difference between directly-made splint group and indirectly-made splint one (p>0.05). 3. Teeth with vital pulps showed higher signal values than teeth with nonvital pulps (p<0.01), and the flow photographs showed heartbeat-synchronous fluctuations and vasomotions, while those were absent in non vital tooth.

• 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.