• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• Economic and Environmental Geology
• /
• v.43 no.3
• /
• pp.235-248
• /
• 2010

To construct the standard methods for evaluation of physicochemical characteristics of tailings in Korea, specific gravity, paste pH, grain size, mineral compositions and heavy metal concentrations of total 26 tailings from 21 metallic mines were analyzed. Specific gravity of tailings ranged from 2.61 to 4.31 (avg. 3.04), and sand and silt grain were dominant in the tailings. Ranges of paste pH were 2.1-9.5 in tailings (7.1-9.2 at magmatic, skarn and hydrothermal replacement deposits and 2.1-9.5 at hydrothermal vein deposits). Additionally, hydrothermal vein deposits could be reclassified into three categories: (1) paste pH>7.0, (2) 4.0

• The Korean Journal of Nuclear Medicine Technology
• /
• v.15 no.2
• /
• pp.104-110
• /
• 2011

Purpose: The object evaluation method about medical institutes of these days increases credibility of consumers about medical services by conducting a certification system about medical institutes. In addition, as nuclear medicine test rooms and diagnosis test medicine room adopt many kinds of international certification systems, the matters regarding safety management of test rooms are being regarded as important. Since the blood test rooms of nuclear medicine are exposed to many harmful factors such as infection from clinical specimen and radioactive isotope reagent, there is a need to pay lots of attention to the safety management of staff and patients. Therefore, this study discusses safety management activities of staff and patients, which are conducted in the blood test rooms of the nuclear medicine department in Asan Medical Center. Materials and Methods: In the blood test rooms of the nuclear medicine department in Asan Medical Center, the matters regarding comprehensive safety management by the person in charge of safety management are offered and all staff members of the test rooms apply them into work. Safety management education is regularly conducted according to established regulations, and infection is prevented through implementation of wearing personal protectors and hand sanitation during test work. In addition, technical safety guides and accident guides for interruption of electric power are provided against emergencies. Through infection management guides, infection prevention and preparation methods for infection are learned and radioactive isotope management, safety management about reagent use and safety guides about harmful chemical substances are being applied to work. Results: The blood test rooms of the nuclear medicine department apply safety management regulations to work. Under the situation where hand sanitation should be conducted, hands are washed to prevent infection between staff and patients, and for preventing infection from clinical specimen, personal protectors are worn. The reagent, which is classified as harmful substance, is separately stored to be easily recognized, radioactive wastes and general medical wastes are also safely managed. Through these lots of safety management activities, safety management awareness of staff members is enhanced, and patients are protected from many dangers. Conclusion: Staff members of the blood test rooms of the nuclear medicine department should be fully aware of safety management regulations and apply them to work. When better safety management suggestions are made, they need to be examined and applied for increasing quality of safety management for staff and patients.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.4
• /
• pp.81-94
• /
• 2006

The water quality impacts of forest fragmentation at the Doam-dam watershed were evaluated in this study. For this ends, the watershed scale model, Soil and Water Assessment Tool (SWAT) model was utilized. To exclude the effects of different magnitude and patterns in weather, the same weather data of 1985 was used because of significant differences in precipitation in year 1985 and 2000. The water quality impacts of forest fragmentation were analyzed temporarily and spatially because of its nature. The flow rates for Winter and Spring has increased with forest fragmentations by $8,366m^3/month$ and $72,763m^3/month$ in the S1 subwatershed, experiencing the most forest fragmentation within the Doam-dam watershed. For Summer and Fall, the flow rate has increased by $149,901m^3/month$ and $107,109m^3/month$, respectively. It is believed that increased flow rates contributed significant amounts of soil erosion and diffused nonpoint source pollutants into the receiving water bodies. With the forest fragmentation in the S1 watershed, the average sediment concentration values for Winter and Spring increased by 5.448mg/L and 13.354mg/L, respectively. It is believed that the agricultural area, which were forest before the forest fragmentation, are responsible for increased soil erosion and sediment yield during the spring thaw and snow melts. For Spring and Fall, the sediment concentration values increased by 20.680mg/L and 24.680mg/L, respectively. Compared with Winter and Spring, the increased precipitation during Summer and Fall contributed more soil erosion and increased sediment concentration value in the stream. Based on the results obtained from the analysis performed in this study, the stream flow and sediment concentration values has increased with forest fragmentation within the S1 subwatershed. These increased flow and soil erosion could contribute the eutrophication in the receiving water bodies. This results show that natural functionalities of the forest, such as flood control, soil erosion protection, and water quality improvement, can be easily lost with on-going forest fragmentation within the watershed. Thus, the minimize the negative impacts of forest fragmentation, comprehensive land use planning at watershed scale needs to be developed and implemented based on the results obtained in this research.

• Journal of Korean Home Economics Education Association
• /
• v.27 no.1
• /
• pp.51-65
• /
• 2015

The purpose of the study was to describe the operating status of "Logic and Writing in Home Economics Education(HEE)" course and the students' perception about the course. For this study, the syllabuses of "Logic and Writing" course in HE education department were reviewed. The survey method was used in this descriptive study. Respondents in this study were 116 students taking the "Logic and Writing in Home Economics Education" course. Questionnaires from the students were collected through on-line and off-line mail. The results of the study were as the follows: First, goals of the "Logic and Writing in HE Education" course were as the follows in order: to enhance writing skill about various subjects related to Home Economics; to foster HE teachers' professionalism through persuasive power logically; to foster writing skill logically; to prepare HE teacher recruitment exam; and to foster teaching skills to enhance students' writing. The student evaluation methods in the course were mainly attendance, examination, tasks, discussion and presentation, and class participation. Second, degree of satisfaction of students taking the course was higher than average. The student respondents perceived that the course was helpful to improve their logical thinking and critical analytical skill, writing skill related to HE education and education, ability to express one's opinion clearly and exactly, understanding of the fundamental concept of logic and the structure and method of the essay, comprehensive understanding of HE education, and ability to solve the problems specifically. However, they responded that it was difficult to practice writing, to discuss and debate frequently, to follow uncertain assessment criteria, and to present frequently. They wanted the instructor to give the feedback more frequently.

• Journal of Nutrition and Health
• /
• v.48 no.5
• /
• pp.419-428
• /
• 2015

Purpose: The current study was conducted in order to develop the Korean Healthy Eating Index (KHEI) for assessing adherence to national dietary guidelines and comprehensive diet quality of healthy Korean adults using the 5th Korea National Health and Nutrition Examination Survey (KNHANES) data. Methods: The candidate components of KHEI were selected based on literature reviews, dietary guidelines for Korean adults, 2010 Dietary Reference Intakes for Koreans (2010 KDRI), and objectives of HP 2020. The associations between candidate components and risk of obesity, abdominal obesity, and metabolic syndrome were assessed using the 5th KNHANES data. The expert review process was also performed. Results: Diets that meet the food group recommendations per each energy level receive maximum scores for the 9 adequacy components of the index. Scores for amounts between zero and the standard are prorated linearly. For the three moderation components among the total of five, population probability densities were examined when setting the standards for minimum and maximum scores. Maximum scores for the total of 14 components are 100 points and each component has maximum scores of 5 (fruit intakes excluding juice, fruit intake including juice, vegetable intakes excluding Kimchi and pickles, vegetable intake including Kimchi or pickles, ratio of white meat to read meat, whole grains intake, refined grains intake, and percentages of energy intake from carbohydrate) or 10 points (protein foods intake, milk and dairy food intake, having breakfast, sodium intake, percentages of energy intake from empty calorie foods, and percentages of energy intake from fat). The KHEI is a measure of diet quality as specified by the key diet recommendations of the dietary guidelines and 2010 KDRIs. Conclusion: The KHEI will be used as a tool for monitoring diet quality of the Korean population and subpopulations, evaluation of nutrition interventions and research.

• Journal of Life Science
• /
• v.26 no.11
• /
• pp.1237-1244
• /
• 2016

A comprehensive analysis of the genetic diversity and relationship of the cold-water fishery walleye pollock (Theragra chalcogramma), the most abundant economically important fishery resource in the East sea of Korea, has not been carried out, despite its importance in Korea. The present study assessed the genetic diversity and relationship between five walleye pollock populations (Korean population, Russian population, USA population, and Japanese populations) of T. chalcogramma using eight microsatellite DNA (msDNA) markers to provide the scientific data for the preservation and management of the Pollock fishery resource. The results of the analysis of 186 individuals of the Pollock revealed a range of 7.13-10.63 numbers of alleles (mean number of alleles=9.05). The means of observed heterozygosity ($H_O$), expected heterozygosity ($H_E$) were 0.732 and 0.698, respectively. The results of genetic distance, Pairwise $F_{ST}$, UPGMA (UPGMA: un-weighted pair-group method with an arithmetical average) (the phylogenetic tree), PCA (PCA: Principal Coordinate analysis) analysis pointed to significant differences between the Korean population, Russian population, USA population, and Japanese populations, although small (p<0.05). These results shed light on the genetic diversity and relationships of T. chalcogramma and can be utilized for research on the evaluation and conservation of Korean T. chalcogramma as genetic resources.

• Korean Journal of Biological Psychiatry
• /
• v.3 no.2
• /
• pp.277-287
• /
• 1996

Objective : It has been proposed that cognition and related aspects of mental functioning are decreased in depression as well as in alcoholism. The objective of the study was to compare behavioral side effects of paroxetine and amitriptyline in depressed patients accompanied by alcoholism. The focused comparisons were drug effects concerning psychomotor performance, cognitive function, sleep and daytime sleepiness during the first 2 weeks of treatment. Methods : After an alcohol detoxification period(3 weeks) and a washout period(1 week), a total of 20 male inpatients with alcohol use disorder (DSM-IV), who also had a major depressive episode(DSM-IV), were treated double-blind with paroxetine 20mg/day(n=10) or amitriptyline 25mg/day(n=10) for 2 weeks. All patients were required to have a scare of at least 18 respectively on bath the Hamilton Rating Scale far Depression(HAM-D) and Beck Depression Inventory(BDI) at pre-drug baseline. Patients randomized to paroxetine received active medication in the morning and placebo in the evening whereas those randomized to amitriptyline received active medication in the evening and placebo in the morning. All patients performed the various tasks in a test battery at baseline and at days 3, 7 and 14. The test battery included : critical flicker fusion threshold for sensory information processing capacity : choice reaction time for gross psychomotor performance : tracking accuracy and latency of response to peripheral stimulus as a measure of line sensorimotor co-ordination and divided attention : digit symbol substitution as a measure of sustained attention and concentration. To rate perceived sleep and daytime sleepiness, 10cm line Visual analogue scales were employed at baseline and at days 3, 7 and 14. The subjective rating scales were adapted far this study from Leeds sleep Evaluation Questionnaire and Epworth Sleepiness Scale. In addition a comprehensive side effect assessment, using the UKU side effect rating scale, was carried out at baseline and at days 7 and 14. The efficacy of treatment was evaluated using HAM-D, BDI and clinical global impression far severity and improvement at days 7 and 14. Results : The pattern of results indicated thai paroxetine improved performance an mast of the lest variables and also improved sleep with no effect on daytime sleepiness aver the study period. In contrast, amitriptyline produced disruption of performance on same tests and improved sleep with increased daytime sleepiness in particular at day 3. On the UKU side effect rating scale, mare side effects were registered an amitriptyline. The therapeutic efficacy was observed in favor of paroxetine early in day 7. Conclusion : These results demonstrated thai paroxetine in much better than amitriptyline for the treatment of depressed patients accompained by alcoholism at least in terms of behavioral safety and tolerability, furthermore the results may assist in explaining the therapeutic outcome of paroxetine. For example, and earlier onset of antidepressant action of paroxetine may be caused by early improved cognitive function or by contributing to good compliance with treatment.

• Research in Community and Public Health Nursing
• /
• v.11 no.1
• /
• pp.289-316
• /
• 2000

우리나라의 경우 지역사회 재난 관리계획과 훈련이 보건의료적 모형이라기 보다는 민방위 모형에 입각하기 때문에 사고 현장에서의 환자 중증도 분류, 합리적 환자배분 및 이송, 병원 응급실에서의 대처 등이 체계적으로 이루어지지 못하고 있으며, 지역사회가 이에 즉각적으로 반응할 수 없다. 본 연구는 삼풍 붕괴사고 시에 대응방식과 그 후의 우리나라 응급의료 체계를 분석함으로써 대형사고 예방과 재난관리를 위한 우리나라 응급의료체계의 개선방안과 간호교육에서의 준비부분을 제시하고자 한다. 1 삼풍 사고 발생시에는 이를 관장할 만한 법적 근거인 인위적 재해에 관한 재난관리법이 없었다. 따라서 현장에서는 의학적 명령체계를 확보하지 못했기 때문에 현장에서의 응급 처치는 전혀 이루어지지 못하였다. 현장에서의 중증도 분류. 응급조치와 의뢰, 병원과 현장본부 그리고 구급차간의 통신 체계 두절, 환자 운송 중 의료지시를 받을 수 있도록 인력, 장비, 통신 체계가 준비되지 못하였던 점이 주요한 문제였다. 또한 병원 응급실에서는 재난 계획이 없거나 있었더라도 이를 활성화하여 병원의 운영 체계를 변환해가지 못하였다. 2. 삼풍백화점 붕괴사고 한달 후에는 인위적 재해에 대한 재난관리법이 제정되고, 행정부 수준별로 매년 지역요구에 합당한 재난관리 계획을 세우도록 법으로 규정하였다. 재난 관리법에는 보건의료 측면에서의 현장대응, 주민 참여, 응급 의료적 대처, 정보의 배된. 교육/훈련 등이 포함되어 있어야 한다. 그러나 법적 기반이 마련된 이후에도 한국 재난 계획 내에는 응급의료 측면의 대응 영역은 부처간 역할의 명시가 미흡하며, 현장에서의 응급 대응과정을 수행할 수 있는 운영 지침이 없이 명목상 언급으로 그치고 있기 때문에 계획을 활성화시켜 지역사회에서 운영하기는 어렵다. 즉 이 내용 속에는 사고의 확인 /공고, 응급 사고 지령, 요구 평가, 사상자의 중증도 분류와 안정화, 사상자 수집, 현장 처치 생명보존과 내과 외과적 응급처치가 수반된 이송, 사고 후 정신적 스트레스 관리, 사고의 총괄적 평가 부분에 대한 인력간 부처간 역할과 업무가 분명히 제시되어 있지 못하여, 사고 발생시 가장 중요한 연계적 업무 처리나 부문간 협조를 하기 어렵다. 의료 기관과 응급실/중환자실, 시민 안전을 책임지고 있는 기관들과의 상호 협력의 연계는 부족하다. 즉 현재의 재난 대비 계획 속에는 부처별 분명한 업무 분장, 재난 상황에 따른 시나리오적 대비 계획과 이를 훈련할 틀을 확보하고 있지 못하다. 3. 지방 정부 수준의 재난 계획서에는 재난 발생시 보건의료에 관한 사항 전반을 공공 보건소가 핵심적 역할을 하며 재난 관리에 대처해야 된다고 규정하고 있다. 그러므로 보건소는 지역사회 중심의 재난 관리 계획을 구성하고 이를 운영하며, 재난 현장에서의 응급 치료 대응 과정은 구조/ 구명을 책임지고 있는 공공기관인 소방서와 지역의 응급의료병원에게 위임한다. 즉 지역사회 재난 관리 계획이 보건소 주도하에 관내 병원과 관련기관(소방서. 경찰서)이 협동하여 만들고 업무를 명확히 분담하여 연계방안을 만든다. 이는 재난관리 대처에 성공여부를 결정하는 주요 요인이다. 4 대한 적십자사의 지역사회 주민에 대한 교육 프로그램은 연중 열리고 있다. 그러나 대부분의 교육주제는 건강증진 영역이며. 응급의료 관리는 전체 교육시간의 8%를 차지하며 이중 재난 준비를 위한 주민 교육 프로그램은 없다. 또한 특정 연령층이 모여있는 학교의 경우도 정규 보건교육 시간이 없기 때문에 생명구조나 응급처치를 체계적으로 배우고 연습할 기회가 없으면서 국민의 재난 준비의 기반확대가 되고 있지 못하다. 5. 병원은 재난 관리 위원회를 군성하여 병원의 진료권역 내에 있는 여러 자원을 감안한 포괄적인 재난관리계획을 세우고, 지역사회를 포함한 훈련을 해야 한다. 그러나 현재 병원은 명목상의 재난 관리 계획을 갖고 있을 뿐이다. 6. 재난관리 준비도를 평가할 때 병원응급실 치료 팀의 인력과 장비 등은 비교적 기준을 충족시키고 있었으나 병원의 재난 관리 계획은 전혀 훈련되고 있지 못하였다 그러므로 우리나라 재난 관리의 준비를 위해서는 현장의 응급의료체계, 재난 대응 계획, 이의 훈련을 통한 주민교육이 선행되어야만 개선될 수 있다. 즉 민방위 훈련 모델이 아닌 응급의료 서비스 모델에 입각한 장기적 노력과 재원의 투입이 필요하며, 지역사회를 중심으로 대응 준비와 이의 활성화 전략 개발, 훈련과 연습. 교육에 노력을 부여해야 한다. 7. 현장의 1차 응급처치자에 대해서는 법적으로 명시하고 있는 역할이 없다. 한국에서는 응급구조사 1급과 2급에 대한 교육과 규정을 1995년 이후 응급의료에 관한 법률에서 정하고 있다. 이 교육과정은 미국이 정하고 있는 응급구조사 과정 기준과 유사하지만 실습실이나 현장에서의 실습시간이 절대적으로 부족하다. 덧붙여 승인된 응급구조사 교육 기관의 강사는 강사로서의 자격기준을 충족할 뿐 아니라 실습강사는 대체적으로 1주일의 1/2은 응급 구조차를 탑승하여 현장 활동을 끊임없이 하고 있으며, 실습은 시나리오 유형으로 진행된다. 그러므로 우리나라의 경우 응급 구조사가 현장 기술 인력으로 역할 할 수 있도록 교과과정 내에서 실습을 강화 시켜야하며, 졸업생은 인턴쉽을 통한 현장 능력을 배양시키는 것이 필요하다. 8. 간호사의 경우 응급전문간호사의 자격을 부여받게 됨에 따라, 이를 위한 표준 교육 지침을 개발함으로써 병원 전 처치와 재난시 대응할 수 있는 역량을 보완해야 한다. 또한 현 자격 부여 프로그램 내용을 고려하여 정규자격 간호사가 현장 1차 치료자(first responder)로 역할 할 수 있도록 간호학 교과과정을 부분 보완해야한다.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.39 no.3
• /
• pp.98-106
• /
• 2011

This study was performed with the purpose of providing basic data for the improvement of zoning regulations of urban nature park by analyzing the present problem which occurred during last 6 years from the year of 2005 when the program was introduced for the first time. The study was processed first by the analysis of the cases of problems evoked by citizens, second the other problems was delineated by interviews of officials, at last the validity of all of the problems was verified by a group of professionals through delphi method. The results can be summarized as follows. 1. In relation to the designation and management of urban national park areas, designation criteria, designation process, maintenance, laws and regulations and 20 other items were found to be problematic. After Delphi method, 5 items were removed and 2 added. The results showed that there were 17 problematic items in total. 2. Regarding the problem of criteria for designation, which are, lack of priority(weights), lack of objectivity due to the difficulty to use quantitative evaluation method, incompatibility for contaminated land environmental impact assessment, incompatibility of land suitability assessment, lack of detailed field survey standards, lack of national park area standards, and 6 other items. 3. Regarding the problem of designation process, which are, the occurrence of civil appeals for designating a new national park, the needs of feasibility study on the urban national park areas constructed before urban national park guidelines came out, lack of a comprehensive review of the boundaries set when determining national park area management plan, poor temporal and financial conditions for an accurate field survey, and 4 other items. 4. Regarding the problem of maintenance management, which are, lack of management system in each space, lack of effectiveness of Urban Nature Park Area Management Plan among master plans for park and green areas, the occurrence of dual managers due to dual natures such as purpose area and city park, lack of professional manpower to manage park areas, and 4 other items. 5. Regarding the problems of regulation guidelines, which are, lack of separate urban park area management plan, incompatibility of the permitted facilities in the park to the park area standards, lack of feasibility study on urban park areas, and 3 other items.