• Journal of Preventive Medicine and Public Health
• /
• 제57권1호
• /
• pp.28-36
• /
• 2024

Objectives: This study investigated the impact of socioeconomic factors and sexual orientation-related attributes on the rates of coronavirus disease 2019 (COVID-19) vaccination and infection among men who have sex with men (MSM). Methods: A web-based survey, supported by the National Research Foundation of Korea, was conducted among paying members of the leading online portal for the lesbian, gay, bisexual, transgender, or queer and questioning (LGBTQ+) community in Korea. The study participants were MSM living in Korea (n=942). COVID-19 vaccination and infection were considered dependent variables, while sexual orientation-related characteristics and adherence to non-pharmacological intervention (NPI) practices served as primary independent variables. To ensure analytical precision, nested logistic regression analyses were employed. These were further refined by dividing respondents into 4 categories based on sexual orientation and disclosure (or "coming-out") status. Results: Among MSM, no definitive association was found between COVID-19 vaccination status and factors such as socioeconomic or sexual orientation-related attributes (with the latter including human immunodeficiency virus [HIV] status, sexual orientation, and disclosure experience). However, key determinants influencing COVID-19 infection were identified. Notably, people living with HIV (PLWH) exhibited a statistically significant predisposition towards COVID-19 infection. Furthermore, greater adherence to NPI practices among MSM corresponded to a lower likelihood of COVID-19 infection. Conclusions: This study underscores the high susceptibility to COVID-19 among PLWH within the LGBTQ+ community relative to their healthy MSM counterparts. Consequently, it is crucial to advocate for tailored preventive strategies, including robust NPIs, to protect these at-risk groups. Such measures are essential in reducing the disparities that may emerge in a post-COVID-19 environment.

• Korean Journal of Radiology
• /
• 제21권1호
• /
• pp.108-116
• /
• 2020

Objective: To investigate the concordance of three international guidelines: the Korean Thyroid Association/Korean Society of Thyroid Radiology, American Thyroid Association, and American College of Radiology for thyroid nodules classified by ultrasonography (US) and the diagnostic performance of simulated size criteria for malignant biopsies. Materials and Methods: A total of 2586 thyroid nodules (≥ 1 cm) were collected from two multicenter study datasets. The classifications of the thyroid nodules were based on three different guidelines according to US categories for malignancy risk, and the concordance rate between the different guidelines was calculated for the classified nodules. In addition, the diagnostic performance of criteria related to four different simulated biopsy sizes was evaluated. Results: The concordance rate of nodules classified as high- or intermediate-suspicion was high (84.1-100%), but low-suspicion or mildly-suspicious nodules exhibited relatively low concordance (63.8-83.8%) between the three guidelines. The differences in sensitivity, specificity, and accuracy between the guidelines were 0.7-19.8%, 0-40.9%, and 0.1-30.5%, respectively, when the original biopsy criteria were applied. The differences decreased to 0-5.9%, 0-10.9%, and 0.1-8.2%, respectively, when simulated, similar biopsy size criteria were applied. The unnecessary biopsy rate calculated with the original criteria (0-33.8%), decreased with the simulated biopsy size criteria (0-8.7%). Conclusion: We found a high concordance between the three guidelines for high- or intermediate-suspicion nodules, and the diagnostic performance of the biopsy criteria was approximately equivalent for each simulated size criterion. The difference in diagnostic performance between the three guidelines is mostly influenced by the various size thresholds for biopsies.

• Korean Journal of Radiology
• /
• 제20권1호
• /
• pp.148-157
• /
• 2019

Objective: To evaluate the current status of managing adverse reactions to iodinated contrast media (ICM) for computed tomography in referral hospitals in South Korea compared with hospitals in other countries. Materials and Methods: This survey investigation involved 59 Korean and 15 overseas hospitals using guideline-based questionnaires consisting of 24 items in 7 main categories related to managing adverse reactions to ICM. Results: Informed written consent with risk factor evaluation was appropriately performed in most of the Korean hospitals. There was considerable variability in assessing renal function across the hospitals; serum creatinine level was used as a reference in 76.4% of Korean hospitals. The Korean hospitals preferred a more stringent approach to determining normal renal function (p = 0.01), withholding metformin (p = 0.01), and fasting before ICM exposure (p < 0.001) compared with overseas hospitals. All the Korean hospitals had an emergency protocol and in-hospital system for adverse reactions to ICM. The Korean (87.7%) and overseas hospitals (100%) were similarly equipped with epinephrine (p = 0.332), but only 38.6% of Korean hospitals were equipped with a bronchodilator (p = 0.004). For patients with a previous hypersensitivity reaction to ICM, 62.3% of Korean hospitals pre-medicated with anti-histamine and corticosteroid according to the severity of the previous reaction, and changed the culprit ICM in 52.8%, while skin test was performed in 17%. Conclusion: In general, Korean referral hospitals were well-prepared regarding informed consent, protocol, and an in-hospital system for managing adverse reactions to ICM. Nevertheless, there was considerable variability in details and management, thus requiring standardization by reflecting current guidelines.

• Annals of Occupational and Environmental Medicine
• /
• 제35권
• /
• pp.26.1-26.15
• /
• 2023

Background: The objective of this study is to investigate the differences in incidence rates of targeted diseases by classification of occupations among construction workers in Korea. Methods: In a subject-based cohort of the Korean Construction Worker's Cohort, we surveyed a total of 1,027 construction workers. As occupational exposure, the classification of occupations was developed using two axes: construction business and job type. To analyze disease incidence, we linked survey data with National Health Insurance Service data. Eleven target disease categories with high prevalence or estimated work-relatedness among construction workers were evaluated in our study. The average incidence rates were calculated as cases per 1,000 person-years (PY). Results: Injury, poisoning, and certain other consequences of external causes had the highest incidence rate of 344.08 per 1,000 PY, followed by disease of the musculoskeletal system and connective tissue for 208.64 and diseases of the skin and subcutaneous tissue for 197.87 in our cohort. We especially found that chronic obstructive pulmonary disease was more common in construction painters, civil engineering welders, and civil engineering frame mold carpenters, asthma in construction painters, landscape, and construction water proofers, interstitial lung diseases in construction water proofers. Conclusions: This is the first study to systematically classify complex construction occupations in order to analyze occupational diseases in Korean construction workers. There were differences in disease incidences among construction workers based on the classification of occupations. It is necessary to develop customized occupational safety and health policies for high-risk occupations for each disease in the construction industry.

• 대한공간정보학회지
• /
• 제22권4호
• /
• pp.21-30
• /
• 2014

최근 소셜 미디어의 활성화로 인해 소셜 네트워크상에서 수많은 자발적 추종자들을 확보한 새로운 형태의 유력자가 대두되고 있다. 소셜 네트워크상에서의 유력자를 탐색하는 연구들이 진행되어 왔고, 관련 서비스가 제공 중에 있으나 이들은 유력자 규명에 있어 위치기반 소셜 네트워크 서비스(LBSNS)가 가지고 있는 위치 정보에 대한 반영이 부족하다는 한계점을 가지고 있었다. 이에 본 연구에서는 공간통계분석기법을 이용하여 LBSNS 데이터를 대상으로 다양한 사회문화적 이슈에 대한 발언에 영향력을 가지는 유력자를 공간적으로 탐색하고, 이를 활용하는 방안을 제시하고자 하였다. 이를 위해 트위터의 지오태깅된 메시지를 분석 데이터로 사용하였으며, 서울시를 공간적 범위로 하여 한 달 동안 총 168,040건의 메시지를 수집하였다. 또한 '정치', '경제', 'IT'를 연구 대상 범주로 설정하고, 데이터 수집 기간 동안 이슈가 되었던 키워드들을 주어진 범주별로 분류하였다. 이를 바탕으로 키워드에 대한 유력자를 파악하기 위한 노출도를 도출하고, 이에 대해 서울시의 행정동을 기준으로 공간결합연산을 실시함으로써 각 키워드에 대한 행정동별 노출도를 산출하였다. 그리고 행정동별로 산출된 노출도의 공간적 의존성을 고려하여 유력지수를 도출하였으며, 키워드별로 상위의 유력지수를 보이는 지역을 유력지역으로 추출하여 이들의 공간적인 분포 특성과 키워드들 간의 공간적 상관성을 분석하였다. 실험 결과, 동일 범주 내에서 키워드 간의 공간적 상관계수는 0.3 이상으로 높은 상관성을 보였으며, 정치범주와 경제범주의 키워드 간 상관계수 역시 평균 0.3으로 비교적 높은 상관성을 보인 반면, 정치범주와 IT범주, 경제범주와 IT범주 키워드 간의 상관계수는 각각 0.18, 0.15로 낮은 상관성을 보였다. 본 연구는 유력자에 대한 연구를 공간 정보의 관점에서 구체화시켰다는 점에서 의의를 가지며, 향후에 gCRM(geographic Customer Relationship Management) 등의 분야에 유용하게 활용될 수 있을 것이다.

• Journal of Nutrition and Health
• /
• 제55권2호
• /
• pp.296-308
• /
• 2022

본 연구는 어머니의 모델링이 유아의 식행동에 미치는 영향을 알아보고자 아이와 어머니의 식사에 대한 행동을 조사하였다. 식행동검사는 아동 영역 58문항, 어머니 영역 20문항으로 구성되어 있으며, 2018년 4월부터 6월까지 어린이집 및 유치원 64개소, 총 1,103명을 조사하였다. 어머니의 식행동 및 식사 지도에 따른 자녀의 식행동을 살펴본 결과, 어머니의 식행동이 불규칙한 경우 자녀의 접근회피성, 과활동성, 불규칙성 점수가 유의하게 높았다. 또한 어머니가 까다로운 경우 자녀의 접근회피성과 감각예민성 점수가 유의하게 높았으며, 자기방임적인 식행동을 보이는 어머니에서 자녀의 과활동성 점수가 유의하게 높게 나타났다. 식사 시 관여도가 높을수록 자녀의 과활동성과 불규칙성 점수가 유의하게 낮았고, 식사 시 규제와 강요가 과할수록 자녀의 과활동성 점수가 유의하게 높은 것으로 나타났다. 위계적 회귀분석을 통해 어머니의 모델링과 식사 지도 유형이 자녀의 식행동에 미치는 영향을 확인한 결과, 자녀의 접근회피성과 감각예민성에는 어머니의 까다로운 식행동과 규제·강요의 식사지도방식이 정적인 영향을 미치며, 자녀의 접근회피성은 어머니의 식사지도방식보다는 까다로운 식행동에 의해 더 큰 영향을 받는 것으로 나타났다. 자녀의 과활동성에는 어머니의 불규칙하고 자기방임적인 식행동과 관심과 규제가 많은 식사 지도 방식이 정적인 영향을 미치며, 어머니의 모델링보다는 식사지도방식에 의해 영향을 많이 받는 것으로 조사되었다. 자녀의 불규칙성에는 어머니의 불규칙하고 까다로운 식행동과 과도한 관심과 규제가 많은 식사 지도 방식이 정적인 영향을 미치며, 식사 지도 방법보다 어머니의 불규칙하고 까다로운 식행동이 더 큰 영향을 미치고 있는 것으로 나타났다. 본 결과를 통해 유아의 올바른 성장과 식생활 형성을 위해서는 자녀가 어머니의 식행동을 모델링할 수 있도록 어머니의 자기 관리적이고 규칙적인 식생활을 통해 자녀들에게 모범을 보일 수 있도록 해야하며, 유아의 영양교육을 계획할 경우 반드시 어머니 교육을 병행하고 아이들의 식행동과 어머니의 식행동을 함께 고려한 맞춤형 솔루션 제공이 중요함을 확인하였다.

• 한국식품조리과학회지
• /
• 제26권5호
• /
• pp.603-616
• /
• 2010

최근 외식업소에서 식중독 발생이 급증함에 따라 이에 대한 대책이 절실히 필요함이 제기되었으나, 생산과 소비가 동시에 일어나는 외식업체의 특성상 위생 감시만으로는 통제가 어려워 업소 내의 자주적인 위생관리 방안이 마련되어야 할 실정이다. 본 연구에서는 외식업체의 위생관리 수준의 향상을 목표로 외식업체의 자율적인 위생관리를 위한 식중독 발생 위험요인과 식품위생법 조항을 포괄하는 모니터링 도구를 개발하고, 이를 이용해 외식업소의 식중독 발생 위험요인과 법적기준의 준수율을 조사하였다. 본 연구의 조사 결과를 요약하면 다음과 같다. 첫째, 대상 외식업소의 전체 준수율의 평균은 68.6%였고, 영역별 준수율은 서류비치 영역이 75.6%로 가장 높았고, 시설 설비관리 영역이 74.5%, 생산단계별 위생관리 영역이 63.9%로 뒤를 이었으며 개인위생관리 영역이 63.0%로 가장 낮은 준수율을 보였다. 둘째, 운영 형태에 따라 다점포 체인업소는 85.5%, 단독 경영업소는 51.6%의 전체 준수율을 보여 p<0.001 범위의 매우 유의적인 결과를 보였다. 하지만 각각의 운영 형태내에서 업종(한식, 중식, 일식, 양식)간에 따른 차이는 전체 준수율과 영역 수준 모두에서 나타나지 않았다. 셋째, CDC에서 규명한 식중독 발생 주요 5대 요인에 기여할 수 있는 인자의 항목 25개 중 준수율이 50% 이하로 집중적인 주의와 관리가 필요한 식중독 발생 잠재 위험요인에는 시설?설비 영역에 '조리장, 창고, 냉장 냉동고의 문서화된 청소계획을 가지고 있는가?'(48.3%), '사용하기 편리한 곳에 전용 수세시설을 구비하고 있는가?'(35.0%)가 해당되었고, 개인위생 관리 영역에서는 '종업원 위생교육을 실시하고, 평가기록을 유지하는가?'(50.0%), '손 세척을 올바르게 실행하는가?'(42.3%)의 항목이 포함 되었다. 생산단계별 위생 관리 영역에서는 '검수 시 식품의 품질과 온도 및 포장상태를 확인하고 기록하는가?'(35.0%), '재가열 온도를 준수하는가?'(44.4%), '냉동식품을 올바르게 해동하고, 해동 후 즉시 사용하는가?'(32.8%), '열장 또는 냉장 보관하거나 조리완료 1시간 반 이내에제공하는가?'(50.0%), '적정 배식 도구를 사용하는가?'(44.8%)의 항목이 나타났다. 서류 비치 영역에서 50% 미만의 준수율을 보인 항목은 존재하지 않았다. 넷째, 국내 식품위생법에서 정하고 있는 기준을 나타낸항목은 시설 설비관리 영역에 11항목, 개인위생관리영역에 3항목, 생산단계별 위생관리 영역에 6항목, 문서 영역에 3항목이 포함되었고, 그 준수율은 35.0-100%로 다양하게 나타났다. 운영 형태에 따른 준수율 비교 결과, 다점포 체인업소와 단독 경영업소 간에 있어 모든 영역에서 매우 유의적인 차이를 나타내어 단독 경영업소의 위생관리 향상을 위한 노력이 많이 요구되었고, 모든 대상 업소에서 식중독 발생과 큰 연관성이 있다고 규명된 요인의 항목 중 개인위생관리와 생산단계별 위생관리 영역에서의 준수율이 많이 미흡하여 위생교육 및 훈련을 통해 손 씻기의 습관화와 검수, 해동, (재)가열, 조리 후 보관의 매 단계에서 시간과 온도관리가 보다 철저히 이루어져야 하겠다. 국내 식품위생법의 준수율의 범위가 다양하게 나타났는데, 그 중 식중독 발생과 관련성이 큰 항목이 낮은 준수율을 보여 이에 대한 법적 감시와 규제를 강화되어야 할 것으로 사료된다. 또한 무엇보다 다양한 교육매체를 개발, 제공하여 식품취급자들의 인식변화와 행동수정을 유도하는 것이 외식업소의 위생관리 수준을 한 단계 높이는 데 필수적이라 하겠다.

• 보건교육건강증진학회지
• /
• 제18권3호
• /
• pp.11-36
• /
• 2001

The purpose of this study is to analyze health behavior by comparing the difference between self-perceived health status and health examination results. The study subjects consist of 7,702 people aged over 20, surveyed by Health Interview survey, Health Examination survey, Dietary Life survey, Health Consciousness and Behavior survey. Data used in the study are drawn from raw data from a 1998 National Health and Nutrition survey. General characteristics variables are sex, age, education level, residential area, marital status, occupation, and living standard while dichotomous variables, ‘not healthy’ and ‘healthy’ are used to measure self-perceived health status. Variables for health examination results are high blood pressure, high cholesterol, diabetes, liver diseases, liver inflammation, kidney diseases, normal weight, regular diet, optimum sleeping time(7-8 hours), regular health examination and health behavior practice group. Major findings of the study are as follows: 1) Analysis of self-perceived health status and health behavior by disease: Variables significantly correlated with high self-perceived health status have strong associations with high health behavior practice, which supports the hypothesis that as one has high self-perceived health status, one is more likely to practice health promoting behavior. The results of analysis of health behavior differences by dividing subjects into two categories, ‘cases of illness’ and ‘cases of no illness’ indicate that drinking, sleeping time, health examination are significant variables (p〈0.001, 0.05) whereas smoking, weight control, regular exercise, regular diet are not significant. 2) Analysis of disparity patterns between self-perceived health status and health examination: The hypothesis that health behaviors would be different according to the disparity pattern between self-perceived health status and health examination is supported as a result of χ2 test. Among Type I : Self-perceived health status is high and actual health status is good (no disease) Type II: Self-perceived health status is high and actual health status is poor(have disease) Type III: Self-perceived health status is low and actual health status is good(no disease) Type IN: Self-perceived health status is low and actual health status is poor(have disease) Type I and Type IV show no disparity, Type I shows the highest health promoting behavior whereas Type IV shows the lowest health promoting behavior. Type II, and III, compared to Type I, practise lower health promoting behavior. Multi-logistics regression analysis was conducted to find out the degree of impact on health behavior. Independent variables are general characteristics, self-perceived health status and health examination result and presence of illness, while the dependent variable is health promoting behavior. The analysis of the impact of self-perceived health status on the health promoting behavior shows that smoking, drinking, weight control, regular exercise, health examination practice, and/or regular diet are significantly correlated to self-perceived health status. High self-perceived health status is inversely related to high health promoting behavior. This finding supports the hypothesis that the higher one perceives one's health, the more likely one is to practice health promoting behavior. On the contrary, the presence of illness has little impact on health promoting behavior. 3) Multiple logistics analysis on how disparity patterns between self-perceived health status and health examination affect health behavior: The results of multiple logistics analysis made on health behavior variables compared to the standard variable are as follows: When analyzed on the standard of Type I, smoking is a significant risk factor for the Type IV. In case of drinking, all the patterns show a high probability of relative risk ratio. With regard to weight control, it is a risk factor for Type II while all the patterns show high probability of not practising when analyzed on the standard of type IV. Type III and IV show high probability of not doing regular exercise while Type IV, shows a high probability of not taking appropriate sleeping time. When analyzed on the standard of type IV, all the patterns show a high probability of not taking health examinations. Type III and IV show a high probability of not having regular meals. As for overall health promoting behavior, Type III and IV show a high relative risk ratio. These two groups have low self-perceived health status. It implies that self-perceived health status has significant impact on health promoting behavior. This is also supported by the fact that Type I with high self-perceived health status and no illness shows a high practice rate of health promoting behavior. Types II and III the groups with high disparity between self-perceived health status and health examination results, show a low practice rate of health promoting behavior when compared to Type I. Type IV, that is the group with low self-perceived health status and actual illness, shows the lowest practice of health promoting behavior. It is highly probable that this type proves to be the poorest health group.

• 한국산학기술학회논문지
• /
• 제18권1호
• /
• pp.625-636
• /
• 2017

이상지질혈증, 고혈압, 그리고 당뇨병은 심뇌혈관질환의 주된 위험인자로서, 한국 성인의 이상지질혈증 유병률은 높지만 관리가 미흡하다는 문제점이 제기되어 왔다. 본 연구의 목적은 이상지질혈증에 초점을 두고 이러한 위험요인들의 유병률, 인지율, 치료율, 그리고 조절률을 조사하는 것이다. 이를 위해 한국건강관리협회 강원지부에서 2013년과 2014년에 국가건강검진을 받은 30세 이상의 성인 58,121명(남자 29,123명, 여자 28,998명)의 자료를 분석하였다. 이상지질혈증의 정의는 진단기준이 아닌, 이의 관리를 위한 치료기준을 따랐다. 즉 심뇌혈관질환 위험요인과 수준을 평가하여, 위험군에 따라 차등적인 LDL-콜레스테롤의 치료목표를 적용하였다. 연령 표준화된 유병률은 이상지질혈증이 32.5%로 가장 높았고, 그 다음으로 고혈압 25.1%, 당뇨병 9.4%의 순이었다. 유병자 기준 연령 표준화된 인지율은 고혈압 76.7%, 당뇨병 74.7%이었지만, 이상지질혈증은 10.6%에 불과했다. 유병자의 연령 표준화된 치료율은 고혈압 74.6%, 당뇨병 70.2%이었지만, 이상지질혈증은 9.4%로 가장 낮았다. 치료 받은 사람의 조절률은 고혈압이 75.8%로 가장 높았고, 그 다음으로 이상지질혈증 63.3%, 당뇨병 43.9%의 순이었다. 그리고 심뇌혈관질환 위험수준이 높은 집단일수록, 고LDL-콜레스테롤혈증의 조절률은 더 낮았다. 이상지질혈증의 유병률은 고혈압과 당뇨병보다 더 높았지만, 인지율과 치료율은 매우 낮은 수준이었다. 이러한 결과는 이상지질혈증 유병자들에 대한 관리 수준을 향상시키기 위한 효과적인 방안이 모색되어야 한다는 것을 제시한다. 이를 위해서는 이상지질혈증에 대한 현행 진단 위주의 국가건강검진체계를 치료 중심의 관리체계로 전환하고, 특히 심뇌혈관질환 고위험군에 대한 사후관리를 강화해야 한다.

• Asia pacific journal of information systems
• /
• 제19권2호
• /
• 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.