• 대한산업공학회지
• /
• 제16권2호
• /
• pp.135-147
• /
• 1990

A nuclear power plant can be viewed as a large complex man-machine system where high system reliability is obtained by ensuring that sub-systems are designed to operate at a very high level of performance. The chance of severe accident involving at least partial core-melt is very low but once it happens the consequence is very catastrophic. The prediction of risk in low probability, high-risk incidents must be examined in the contest of general engineering knowledge and operational experience. Engineering knowledge forms part of the prior information that must be quantified and then updated by statistical evidence gathered from operational experience. Recently, Bayesian procedures have been used to estimate rate of accident and to predict future risks. The Bayesian procedure has advantages in that it efficiently incorporates experts opinions and, if properly applied, it adaptively updates the model parameters such as the rate or probability of accidents. But at the same time it has the disadvantages of computational complexity. The predictive distribution for the time to next incident can not always be expected to end up with a nice closed form even with conjugate priors. Thus we often encounter a numerical integration problem with high dimensions to obtain a predictive distribution, which is practically unsolvable for a model that involves many parameters. In order to circumvent this difficulty, we propose a method of approximation that essentially breaks down a problem involving many integrations into several repetitive steps so that each step involves only a small number of integrations.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2014년도 추계학술대회 논문집
• /
• pp.347-349
• /
• 2014

Model verification and validation (V&V) is a current research topic to build computational models with high predictive capability by addressing the general concepts, processes and statistical techniques. The hypothesis test for validity check is one of the model validation techniques and gives a guideline to evaluate the validity of a computational model when limited experimental data only exist due to restricted test resources (e.g., time and budget). The hypothesis test for validity check mainly employ Type I error, the risk of rejecting the valid computational model, for the validity evaluation since quantification of Type II error is not feasible for model validation. However, Type II error, the risk of accepting invalid computational model, should be importantly considered for an engineered products having high risk on predicted results. This paper proposes a technique named as the response-adaptive experimental design to reduce Type II error by adaptively designing experimental conditions for the validation experiment. A tire tread block problem and a numerical example are employed to show the effectiveness of the response-adaptive experimental design for the validity evaluation.

• 지능정보연구
• /
• 제26권2호
• /
• pp.105-129
• /
• 2020

본 연구는 부도위험 예측을 위해 K-IFRS가 본격적으로 적용된 2012년부터 2018년까지의 기업데이터를 이용한다. 부도위험의 학습을 위해, 기존의 대부분 선행연구들이 부도발생 여부를 기준으로 사용했던 것과 다르게, 본 연구에서는 머튼 모형을 토대로 각 기업의 시가총액과 주가 변동성을 이용하여 부도위험을 산정했으며, 이를 통해 기존 방법론의 한계로 지적되어오던 부도사건 희소성에 따른 데이터 불균형 문제와 정상기업 내에서 존재하는 부도위험 차이 반영 문제를 해소할 수 있도록 하였다. 또한, 시장의 평가가 반영된 시가총액 및 주가 변동성을 기반으로 부도위험을 도출하되, 부도위험과 매칭될 입력데이터로는 비상장 기업에서 활용될 수 있는 기업 정보만을 활용하여 학습을 수행함으로써, 포스트 팬데믹 시대에서 주가 정보가 존재하지 않는 비상장 기업에게도 시장의 판단을 모사하여 부도위험을 적절하게 도출할 수 있도록 하였다. 기업의 부도위험 정보가 시장에서 매우 광범위하게 활용되고 있고, 부도위험 차이에 대한 민감도가 높다는 점에서 부도위험 산출 시 안정적이고 신뢰성 높은 평가방법론이 요구된다. 최근 머신러닝을 활용하여 기업의 부도위험을 예측하는 연구가 활발하게 이루어지고 있으나, 대부분 단일 모델을 기반으로 예측을 수행한다는 점에서 필연적인 모델 편향 문제가 존재하고, 이는 실무에서 활용하기 어려운 요인으로 작용하고 있다. 이에, 본 연구에서는 다양한 머신러닝 모델을 서브모델로 하는 스태킹 앙상블 기법을 활용하여 개별 모델이 갖는 편향을 경감시킬 수 있도록 하였다. 이를 통해 부도위험과 다양한 기업정보들 간의 복잡한 비선형적 관계들을 포착할 수 있으며, 산출에 소요되는 시간이 적다는 머신러닝 기반 부도위험 예측모델의 장점을 극대화할 수 있다. 본 연구가 기존 머신러닝 기반 모델의 한계를 극복 및 개선함으로써 실무에서의 활용도를 높일 수 있는 자료로 활용되기를 바라며, 머신러닝 기반 부도위험 예측 모형의 도입 기준 정립 및 정책적 활용에도 기여할 수 있기를 희망한다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권9호
• /
• pp.4091-4094
• /
• 2014

Background: This study used the receiver operating characteristic curve (ROC) to analyze Surveillance, Epidemiology and End Results (SEER) bronchioaveolar carcinoma data to identify predictive models and potential disparity in outcomes. Materials and Methods: Socio-economic, staging and treatment factors were assessed. For the risk modeling, each factor was fitted by a Generalized Linear Model to predict cause specific survival. The area under the ROC was computed. Similar strata were combined to construct the most parsimonious models. A random sampling algorithm was used to estimate modeling errors. Risk of cause specific death was computed for the predictors for comparison. Results: There were 7,309 patients included in this study. The mean follow up time (S.D.) was 24.2 (20) months. Female patients outnumbered male ones 3:2. The mean (S.D.) age was 70.1 (10.6) years. Stage was the most predictive factor of outcome (ROC area of 0.76). After optimization, several strata were fused, with a comparable ROC area of 0.75. There was a 4% additional risk of death associated with lower county family income, African American race, rural residency and lower than 25% county college graduate. Radiotherapy had not been used in 2/3 of patients with stage III disease. Conclusions: There are socio-economic disparities in cause specific survival. Under-use of radiotherapy may have contributed to poor outcome. Improving education, access and rates of radiotherapy use may improve outcome.

• Korean Journal of Radiology
• /
• 제22권3호
• /
• pp.344-353
• /
• 2021

Objective: The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm. Materials and Methods: Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria. Results: At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows: sensitivity 85.7% (95% confidence interval [CI]: 0.834-0.877), specificity 67.5% (95% CI: 0.636-0.712), PPV 82.1% (95% CI: 0.797-0.843), NPV 73.0% (95% CI: 0.691-0.766), and AUC 0.771 (95% CI: 0.750-0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541-0.995), specificity 70.0% (95% CI: 0.354-0.919), PPV 75.0% (95% CI: 0.428-0.933), NPV 87.5% (95% CI: 0.467-0.993), and AUC 0.800 (95% CI: 0.563-0.943). Conclusion: We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.

• 한국인공지능학회지
• /
• 제6권1호
• /
• pp.16-20
• /
• 2018

The development of modern information technology has increased the amount of big data about patients' information and diseases. In this study, we developed a prediction model of diabetes using the health examination data provided by the public data portal in 2016. In addition, we graphically visualized diabetes incidence by sex, age, residence area, and income level. As a result, the incidence of diabetes was different in each residence area and income level, and the probability of accurately predicting male and female was about 65%. In addition, it can be confirmed that the influence of X on male and Y on female is highly to affect diabetes. This predictive model can be used to predict the high-risk patients and low-risk patients of diabetes and to alarm the serious patients, thereby dramatically improving the re-admission rate. Ultimately it will be possible to contribute to improve public health and reduce chronic disease management cost by continuous target selection and management.

• Journal of the Korean Data and Information Science Society
• /
• 제27권5호
• /
• pp.1293-1305
• /
• 2016

본 연구는 한국장학재단 일반상환 학자금 대출 연체자를 대상으로 연체 미회수 그룹으로 분류될 수 있는 위험요인들을 파악하고, 학자금 대출 연체 회수 예측모형을 개발하였다. 또한 개발된 예측모형을 활용하여 그에 따른 신용위험 평점표를 작성하였다. 예측모형 개발은 연체기간에 따라 총 3가지 모형 (Model 1: 연체 1개월 모형, Model 2: 연체 2개월 모형, Model 3: 연체 3개월 이상 모형)으로 로지스틱 회귀분석 분석을 적용하였다. 연체기간 구분은 금융권에서 일반적으로 사용하고 있는 연체회수모형의 단위를 준용하여 1개월 단위를 기준으로 연체 1개월, 연체 2개월, 연체 3개월 이상으로 구분하였다. 연체 1개월 모형 (Model 1)에서는 연체계좌수, 이체일자, 연체잔액, 소득분위가 영향력이 큰 것으로 나타났으며, 연체 2개월 모형 (Model 2)에서는 연체 일수, 연체잔액, 이체일자, 연체금액이 중요한 것으로 나타났다. 마지막으로 연체 3개월 이상 모형 (Model 3)에서는 최근 3개월 이내 연체 횟수, 이체일자, 연체계좌수, 연체액의 영향력이 큰 것으로 나타났다. 본 연구에서 개발된 연체회수 모형이나 평점표를 바탕으로 연체 채권관리를 함에 있어 좀더 세분화된 관리서비스를 제공하고, 상담센터의 상담원이 연체자의 평점에 따라 상담전략을 세울 수 있는 기초자료가 될 수 있을 것으로 사료된다.

• Archives of Plastic Surgery
• /
• 제42권2호
• /
• pp.131-142
• /
• 2015

Literature indicates an increased risk of suicide among women who have had cosmetic breast implants. An explanatory model for this association has not been established. Some studies conclude that women with cosmetic breast implants demonstrate some characteristics that are associated with increased suicide risk while others support that the breast augmentation protects from suicide. A systematic review including data collection from January 1961 up to February 2014 was conducted. The results were incorporated to pre-existing suicide risk models of the general population. A modified suicide risk model was created for the female cosmetic augmentation mammaplasty candidate. A 2-3 times increased suicide risk among women that undergo cosmetic breast augmentation has been identified. Breast augmentation patients show some characteristics that are associated with increased suicide risk. The majority of women reported high postoperative satisfaction. Recent research indicates that the Autoimmune syndrome induced by adjuvants and fibromyalgia syndrome are associated with silicone implantation. A thorough surgical, medical and psycho-social (psychiatric, family, reproductive, and occupational) history should be included in the preoperative assessment of women seeking to undergo cosmetic breast augmentation. Breast augmentation surgery can stimulate a systematic stress response and increase the risk of suicide. Each risk factor of suicide has poor predictive value when considered independently and can result in prediction errors. A clinical management model has been proposed considering the overlapping risk factors of women that undergo cosmetic breast augmentation with suicide.

• 한국학교보건학회지
• /
• 제36권3호
• /
• pp.113-125
• /
• 2023

Purpose: The purpose of this study is to develop a model for predicting mental health risk among adolescents based on health behavior information by employing a hybrid machine learning method. Methods: The study analyzed data of 51,850 domestic middle and high school students from 2022 Youth Health Behavior Survey conducted by the Korea Disease Control and Prevention Agency. Firstly, mental health risk levels (stress perception, suicidal thoughts, suicide attempts, suicide plans, experiences of sadness and despair, loneliness, and generalized anxiety disorder) were classified using the k-mean unsupervised learning technique. Secondly, demographic factors (family economic status, gender, age), academic performance, physical health (body mass index, moderate-intensity exercise, subjective health perception, oral health perception), daily life habits (sleep time, wake-up time, smartphone use time, difficulty recovering from fatigue), eating habits (consumption of high-caffeine drinks, sweet drinks, late-night snacks), violence victimization, and deviance (drinking, smoking experience) data were input to develop a random forest model predicting mental health risk, using logistic and XGBoosting. The model and its prediction performance were compared. Results: First, the subjects were classified into two mental health groups using k-mean unsupervised learning, with the high mental health risk group constituting 26.45% of the total sample (13,712 adolescents). This mental health risk group included most of the adolescents who had made suicide plans (95.1%) or attempted suicide (96.7%). Second, the predictive performance of the random forest model for classifying mental health risk groups significantly outperformed that of the reference model (AUC=.94). Predictors of high importance were 'difficulty recovering from daytime fatigue' and 'subjective health perception'. Conclusion: Based on an understanding of adolescent health behavior information, it is possible to predict the mental health risk levels of adolescents and make interventions in advance.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.323-325
• /
• 2021

In this article, we have improved the prediction of hypertension detection using the feature selection method for the Korean national health data named by the KNHANES database. The study identified a variety of risk factors associated with chronic hypertension. The paper is divided into two modules. The first of these is a data pre-processing step that uses a factor analysis (FA) based feature selection method from the dataset. The next module applies a predictive analysis step to detect and predict hypertension risk prediction. In this study, we compare the mean standard error (MSE), F1-score, and area under the ROC curve (AUC) for each classification model. The test results show that the proposed FIFA-OE-NB algorithm has an MSE, F1-score, and AUC outcomes 0.259, 0.460, and 64.70%, respectively. These results demonstrate that the proposed FIFA-OE method outperforms other models for hypertension risk predictions.