• The Journal of the Korea Contents Association
• /
• v.18 no.8
• /
• pp.525-535
• /
• 2018

If the frequency of a particular class is excessively higher than the frequency of other classes in the classification problem, data imbalance problems occur, which make machine learning distorted. Corporate bankruptcy prediction often suffers from data imbalance problems since the ratio of insolvent companies is generally very low, whereas the ratio of solvent companies is very high. To mitigate these problems, it is required to apply a proper sampling technique. Until now, oversampling techniques which adjust the class distribution of a data set by sampling minor class with replacement have popularly been used. However, they are a risk of overfitting. Under this background, this study proposes ROSE(Random Over Sampling Examples) technique which is proposed by Menardi and Torelli in 2014 for the effective corporate bankruptcy prediction. The ROSE technique creates new learning samples by synthesizing the samples for learning, so it leads to better prediction accuracy of the classifiers while avoiding the risk of overfitting. Specifically, our study proposes to combine the ROSE method with SVM(support vector machine), which is known as the best binary classifier. We applied the proposed method to a real-world bankruptcy prediction case of a Korean major bank, and compared its performance with other sampling techniques. Experimental results showed that ROSE contributed to the improvement of the prediction accuracy of SVM in bankruptcy prediction compared to other techniques, with statistical significance. These results shed a light on the fact that ROSE can be a good alternative for resolving data imbalance problems of the prediction problems in social science area other than bankruptcy prediction.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.24 no.6
• /
• pp.459-471
• /
• 1991

The community structure and micro-scale distribution of plankton in relation to hydrography were investigated in Masan Bay, Korea in October 1989. Warmer and less saline waters with stratification was located in the inner part of the Pudo Strait, and chlorophyll-a and nutrients were higher. Both phytoplankton biomass and nutrients was changed dramatically around the Strait. Offshore/oceanic species in phytoplankton i.e., Chaetoceres decipiens, Rhizosolenia stolterforthii, Rhizosolenia styliformis and Ceratium trichoceros and zooplankton i.e., Sagitta enflata, Oncaea uenusta and Oikopluera longicaudata occurred mainly in well mixed waters of the outer part. This suggests that discontinuous layer seems to play an important role as an approximate border for the plankton population. This layer was located between Station 3 and Station 4 near the Pudo Strait, since the layer consisted of a series of micro-scale discontinuties of salinity and dissolved inorganic nutrients gradient. Phytoplankton patchs of more than 801e1 were found only in the inner part of the bay. Depletion of silicate caused by a rapid assimilation of phytoplankton in the inner part of the bay seemed to be responsible for the decline of blooms.

• 한국벤처창업학회:학술대회논문집
• /
• 2019.04a
• /
• pp.45-49
• /
• 2019

매년 수많은 기업이 폐업하고 있으며 그 숫자만큼 폐업 기업의 대표가 실패하고 있다. 중소기업연구원의 재기성공요인 분석을 통한 중소 벤처기업의 재도전 활성화 연구(백필규,2011)에 의하면 폐업이나 부도의 형태로 실패하는 기업이 급격하게 증가한 시점은 외환위기가 발생했던 1997년과 1998년의 사이로 보고 있다. 부도 기업의 숫자가 외환위기때는 약 4만사, 1999년부터 2010년까지도 평균 4만 7천여사가 발생하였고 폐업기업수는 부도기업보다 훨씬 많아 매년 80~90만개 전후의 기업이 실패하고 있다. 최근 IBK경제연구소의 실패기업인의 재창업 지원제도 설문조사 결과(김나라,2017)에 의하면 최근 5년간 창업기업은 연평균 77만개사, 폐업기업은 69만개사로 조사되었으며 창업기업의 5년 생존율은 27.3%로 OECD 주요회원국 17개국중 최하위를 차지했다. 오랜시간 대한민국 사회는 사업실패는 곧 패가망신이라는 인식이 깊게 자리잡고 있다. 연대보증이나 재기가 어려운 구조로 인하여 폐업기업의 대표가 실패하면 가족들도 모두 힘들어지고 본인은 경제적, 심리적 어려움에 직면하고 있는 현실이다. 정부와 민간 기관들이 실패기업인들의 재기를 지원하기 위하여 다양한 제도를 운영하고 있지만 아직 시행 초기 단계라 보다 재기를 원하는 기업인들에게 맞춤형으로 도움을 줄 수 있는 실질적이고 효과적인 시스템이 구축되었다 할 수 없다. 창업강국인 미국과 중국의 기업가들은 평균 2.8회의 실패 경험을 가지고 있으나 한국의 기업가는 1.3회의 실패 경험을 보유 한다.(중기청, 2014) 폐업기업 대표의 재무적 손실의 규모와 관계없이 불안과 우울이 낮고 회복탄력성이 높은 기업가가 본인과 가족의 삶의 질과 재창업의지에 긍정적인 효과가 있음으로 나타났다. 연구 대상은 재기중소기업개발원의 재기 기업인과 정부 재창업 지원 프로그램에 지원한 재기 기업인을 대상으로 조사 하였으며 본 연구를 통해 폐업기업 대표에게 필요한 사회적 안전장치와 국가의 창업지원이 고용창출, 매출증가로 이어질 수 있는 선순환 구조의 재기 창업 생태계를 만드는 데 일조하고자 한다.

• Proceedings of the Korea Association of Information Systems Conference
• /
• 2000.05a
• /
• pp.115-121
• /
• 2000

본 연구는 인공신경망 모형에서 최적 입력 변수를 선정하기 위하여 새로운 선처리 기법인 약체연결뉴런 제거법을 제안하고 그 예측력의 우월성을 순수 인공신경망과 의사결정트리로 선처리한 인공신경망 모델과 각각 비교했으며, 그 결과를 보면 본 연구에서 제안하고 있는 약체연결뉴런 제거법에 의해 입력변수 선정과정을 거친 모델의 성과가 순수 인공신경망이나 의사결정트리로 선처리한 인공신경망 모델에 비해 예측적중율이 우수한 것으로 나타났다.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.743-745
• /
• 1999

In these days, pumped storaged power plants are under operations mainly for load regulation in power system. After we investigated the governor related electrical circuit of a pumped storaged power plant, we measured speed regulation and dead time of both the governor and the hydraulic system according to the change of variables of the governor during generation. After that the measured speed regulation improved much.

• The Korean Journal of Applied Statistics
• /
• v.26 no.6
• /
• pp.977-985
• /
• 2013

Many situations are classified into more than two categories in real world. In this work, we consider ROC surface and VUS, which are graphical representation methods for classification models with three categories. The standard criteria of AUC for the probability of default based on Basel II is extended to the VUS for ROC surface; therefore, the standardized criteria of VUS for the classification model is proposed. The ranges of AUC, K-S and mean difference statistics corresponding to VUS values for each class of the standard criteria are obtained. The standard criteria of VUS for ROC surface can be established by exploring the relationships of these statistics.

• The Journal of Society for e-Business Studies
• /
• v.22 no.1
• /
• pp.37-49
• /
• 2017

In order to maximize the efficiency and the quality of care in a hospital, hospitals in general operate by appointments. Patient no-shows or missed appointments waste the time of staffs who prepare for appointments, deprive patients of needed care, and increase burden on clinic capacity. The purpose of this study is to examine the importance and significance of factors that affect patient no-shows. The actual data of 50,000 outpatient appointments is compiled and analyzed from a VAMC in the United States with aims to identify the factors that contribute to outpatient no-shows and to draw meaningful implications. Random forest along with logistic regression analysis identify the factors affecting no-shows; appointment leadtime, show-up rate, travel distance, no-show rate in previous period, patient age, severity, complexity of disease, chronic pain, depression, drug dependence. The appointment leadtime, show-up rate, travel distance and previous no-show rate can be improved by enhancing SMS pre-notification and by improving afterwards telephone counseling. For other factors, it needs to study on the service differentiation with the characteristics of each patient group.

• The Korean Journal of Financial Management
• /
• v.15 no.2
• /
• pp.81-105
• /
• 1998

본 연구는 우리나라 주거래은행이 거래기업의 재무곤경감소나 채무조정방법의 선택에 어떤 영향을 미치는 가를 검증해 보기 위한 것이다. 만성적 재무곤경상태에 빠져 있는 52개의 상장 기업을 대상으로 7년간의 누적투자율이나 매출액증가율 및 이익증가율을 조사해 본 결과 주거래은행관계의 척도라 할 수 있는 최대대출비율이나 주식소유 비율이 누적투자율이나 누적매출액증가율에 거의 영향을 미치지 못하는 것으로 드러났다. 그러나 대그룹소속기업들은 재무곤경기간에도 지속적인 투자나 매출액증대를 보여 그룹간의 내부금융이나 신뢰성이 중요한 역할을 하는 것으로 보여진다. 한편 재무곤경비용의 감소를 누적이익증가율이라고 간주한 경우는 주거래은행의 주식보유비율이 누적이익증가율에 (-)의 영향을 미치는 것으로 나타났다. 이는 부도공시기업의 검증결과와도 일치한다. 주거래은행관계의 유효성은 채무조정방법의 선택에서 잘 나타난다. 최대대출비율과 금융기관의 주식소유비율이 높은 기업일수록 사적협상에 성공할 확률이 높은 것으로 나타나 주거래은행을 위시한 주요 채권단들이 채무조정을 주도적으로 이끌어 워크아웃을 성공시킬 가능성이 높으며, 기업자체의 성장성이나 경영지배권 등의 소유구조는 그리 큰 영향을 미치지 못하는 것으로 나타났다.

• The Korean Journal of Financial Management
• /
• v.22 no.2
• /
• pp.213-226
• /
• 2005

Adverse selection is a heavily scrutinized subject within the financial intermediary industry. Consensus is reached regarding its effect on the loan interest rate. Despite the similar features of financial service offered by the credit card, we still have controversy regarding credit card interest rate on how is adverse selection incurred with the change of interest rate. Thus, this paper explores how does the adverse selection, if ever, take place and affect the credit card interest rate. Information asymmetry regarding the credit card users' type represented by the default probability is assumed. The users are assumed to be rational in that they want to minimize the per unit dollar expense associated with the commercial transaction and financing between the two typical payment methods, cash and credit card. Suppliers, i.e. credit card companies, would like to maximize their profit and would be better off with more pervasive use of credit cards over the cash. Then we could show that the increasing credit card interest rate is subject to the adverse selection, sharing the same tenet with that of the bank loan interest rate proposed by Stiglitz and Weiss. Hence the current theory predicts that credit card market also suffers from adverse selection with increasing interest rate.

• Journal of Digital Contents Society
• /
• v.19 no.9
• /
• pp.1751-1758
• /
• 2018

An integrated medical information system that integrates systems consisting of different environments centered on hospital information systems should be provided as a system that prioritizes the improvement of the quality of medical services, customer satisfaction, and patient safety. The RBAC-based medical information system is granted the access right according to task type, role, and rules. Through this, it is possible to use SMS channel, medical reservation and cancellation, customized statistics, and CRM / EMR interworking service using multi-channel to enable communication service without help of counselor and reduce the default rate of reservation patient, Operational improvement services can be extended to medical staff, patients and their families, as well as expanding to important decisions for patients.