• 대한통합의학회지
• /
• 제12권2호
• /
• pp.155-166
• /
• 2024

Purpose : This study aims to identify key factors for predicting dropout risk at the university level and to provide a foundation for policy development aimed at dropout prevention. This study explores the optimal machine learning algorithm by comparing the performance of various algorithms using data on college students' dropout risks. Methods : We collected data on factors influencing dropout risk and propensity were collected from N University. The collected data were applied to several machine learning algorithms, including random forest, decision tree, artificial neural network, logistic regression, support vector machine (SVM), k-nearest neighbor (k-NN) classification, and Naive Bayes. The performance of these models was compared and evaluated, with a focus on predictive validity and the identification of significant dropout factors through the information gain index of machine learning. Results : The binary logistic regression analysis showed that the year of the program, department, grades, and year of entry had a statistically significant effect on the dropout risk. The performance of each machine learning algorithm showed that random forest performed the best. The results showed that the relative importance of the predictor variables was highest for department, age, grade, and residence, in the order of whether or not they matched the school location. Conclusion : Machine learning-based prediction of dropout risk focuses on the early identification of students at risk. The types and causes of dropout crises vary significantly among students. It is important to identify the types and causes of dropout crises so that appropriate actions and support can be taken to remove risk factors and increase protective factors. The relative importance of the factors affecting dropout risk found in this study will help guide educational prescriptions for preventing college student dropout.

• 수산경영론집
• /
• 제42권2호
• /
• pp.1-14
• /
• 2011

We try to find the determinants of fisheries firms' capital structure during the years from 1992 to 2007 in this paper. We also have a comparative analysis of capital raising behavior in pre and post-IMF financial crisis. Regression analysis is used for this empirical study. Dependent variable is leverage ratio and independent variables are firm size, operating risk, proportion of tangible asset, non-debt tax shield effect, sales growth ratio, profitability and dummy variable. We compared the characteristics of fisheries industry with that of manufacturing industy. The determinants of fisheries firms' capital structure and correlation between pre and post-IMF financial crisis are roughly same as the hypothses except a little difference. As a peculiar difference, corrlation between fisheries firms' operating risk and leverage ratio is (+) in the pre-IMF financial crisis, but (-) in the post-IMF financial crisis. Proportion of tangible asset has a (+) correlation with leverage ratio in pre and post-IMF financial crisis, but in case of manufacturing industy, (-) correlation shows in the pre-IMF financial crisis. Because, in the pre-IMF financial crisis, high proportion of tangible asset doesn't play a role of a collateral, but only increase the bankruptcy probability. Non-debt tax shield effect and leverage ratio have (-) correlation in all industry and all period, but only (+) correlation in case of fisheries industry in the pre-IMF financial crisis. Sales growth ratio has no significant relationship with leverage ratio in fisheries industry, and this is not coincide with our hypothsis. We have a limitation of the sample size of fisheries firms and sample period in this study. Further study is required to classify the fisheries industry with in-shore fisheries, deep sea fisheries and cold storage industry.

• 한국콘텐츠학회논문지
• /
• 제22권7호
• /
• pp.395-404
• /
• 2022

본 연구는 1인 가구 중고령자의 비만 영향요인을 교육 수준에 따라 비교분석함으로써, 노인 비만 예방, 그리고 교육 수준에 따른 차별화된 개입방안을 모색하는 데 목적을 두었다. 국민건강영양조사 20년도 자료를 활용하여 총 400명의 응답자의 자료를 통해 교차분석, 로지스틱 회귀분석 등을 실시하였고, 분석 결과 교육 수준이 낮을 경우 비만에 영향을 줄 수 있는 신체활동에서 취약성이 나타났으며, 교육 수준이 높은 집단에서는 외식과 나트륨 섭취 등의 위험요인 점수가 높음을 확인하였다. 교육 수준 구분을 통해 영향요인을 비교분석한 결과, 교육 수준이 낮은 집단에서는 앉아서 보내는 시간이 위험요인이었고, 교육 수준이 높은 집단에서는 나트륨 섭취가 비만 위험요인인 것으로 나타났다. 이를 통해 1인 가구 중고령자들을 대상으로 한 건강한 영양 섭취 및 신체활동 교육을 실시하고, 교육수준에 따라 차별화된 개입 등이 필요함을 논의하였다.

• 핵의학기술
• /
• 제13권3호
• /
• pp.137-142
• /
• 2009

목적: WHO (world health organization)에서는 골밀도뿐만 아니라 대규모 역학연구에서 정리된 골절의 위험인자 분석을 통하여 10년 내 골절위험도(10-year fracture probability)를 실제적으로 임상에 적용시킬 수 있는 소프트웨어 프로그램인 FRAX Tool (fracture risk assessment)이 2008년에 공개되었다. 본 연구는 기존에 사용하고 있는 GE Prodigy사의 골절위험도평가와 $FRAX^{TM}$를 이용한 골절위험도평가를 비교분석하고자 한다. 검사방법: 본원에 골밀도 검사를 시행한 201명($55{\pm}3.5$세)의 여자를 대상으로 GE Prodigy를 이용하여 Femur를 측정하였다. 측정한 Femoral Neck의 BMD (bone mineral density)를 구하여 GE Prodigy의 T-값과 골절위험인자를 고려하지 않은 $FRAX^{TM}$의 T-값를 사용하여 10년 내 대퇴골 골절위험도와 주요한 골다공증성 골절 위험도를 계산하여 SPSS통계프로그램으로 GE Prodigy의 골절위험도 평가와 $FRAX^{TM}$의 골절위험도 평가를 비교분석하였다. 결과: GE Prodigy의 T-값($-0.52{\pm}0.97$)과 $FRAX^{TM}$의 T-값 ($-1.45{\pm}0.81$)은 통계적으로 유의한 차이가 있었으며(p=0.000), GE Prodigy의 주요한 골다공증성 골절위험도($9.15{\pm}3.71$)와 $FRAX^{TM}$의 주요한 골다공증성 골절위험도($4.87{\pm}1.51$)도 또한 통계적으로 유의한 차이가 있었다(p=0.000). 그리고 GE Prodigy의 10년내 대퇴골 골절위험도($1.56{\pm}1.48$)와 $FRAX^{TM}$의 10년 내 대퇴골골절위험도($0.53{\pm}0.61$)도 통계적으로 유의한 차이를 보였다(p=0.000). 결론: GE Prodigy와 $FRAX^{TM}$에서 측정한 골절의 절대위험도는 통계적으로 유의한 차이를 보였다. 특히 GE Prodigy의 T-값, 주요한 골다공증성 골절위험도, 10년 내 대퇴골 골절위험도가 높게 측정되었다. 향후 $FRAX^{TM}$에 대한 평가 및 적용에 관한 연구가 필요할 것으로 생각된다.

• 한국환경독성학회:학술대회논문집
• /
• 한국환경독성학회 2003년도 추계국제학술대회
• /
• pp.169-169
• /
• 2003

In Korea, there is no CRA studies and has not well known CRA and not well established their methodologies. Therefore, objectives of this study is to establish the framework of CRA consisting of health risk, economic risk and perceived risk and the detail methodologies of three main component of estimating and comparing those risks for on the three environmental problems of air pollution, indoor air pollution and drinking water contamination which being subjective to the eight sub-problems of hazardous ai. pollutants (HAPs), regulated pollutants (representative as PM10) and Dioxins (PCDDS/ PCDFs) in air pollution, and indoor ai. pollutants (IAPs) and Radon in indoor air pollution, and drinking water pollutants (DWPs), disinfection-by- products(DBPs) and radionuclides in drinking water contamination in Seoul, Korea. And then, their problems set priorities by individual and integrated risk. As a results, ranking of health risk were the following order of indoor air pollution, air pollution and then drinking water contamination, in three environmental problems and of radon, PM10, IAPs, HAPs, DWPs, Dioxins, DBPs, and then radionuclides in eight sub-problems. And that of economic risk were the same order. In the contrary, ranking of perceived risk were the following order of air pollution, drinking water contamination, and then indoor air pollution, and of HAPs, Dioxins, radionuclides, PM10, DWPs, IAPs, Radon and then DBPs.

• 대한간호학회지
• /
• 제33권2호
• /
• pp.162-169
• /
• 2003

Purpose: This study was to compare the predictive validity of Norton Scale(1962), Cubbin & Jackson Scale(1991), and Song & Choi Scale(1991). Method: Data were collected three times per week from 48~72hours after admission based on the four pressure sore risk assessment scales and a skin assessment tool for pressure sore on 112 intensive care unit(ICU) patients in a educational hospital Ulsan during Dec, 11, 2000 to Feb, 10, 2001. Four indices of validity and area under the curve(AUC) of receiver operating characteristic(ROC) were calculated. Result: Based on the cut off point presented by the developer, sensitivity, specificity, positive predictive value, negative predictive value were as follows : Norton Scale : 97%, 18%, 35%, 93% respectively; Cubbin & Jackson Scale : 89%, 61%, 51%, 92%, respectively; and Song & Choi Scale : 100%, 18%, 36%, 100% respectively. Area under the curves(AUC) of receiver operating characteristic(ROC) were Norton Scale .737, Cubbin & Jackson Scale .826, Song & Choi Scale .683. Conclusion: The Cubbin & Jackson Scale was found to be the most valid pressure sore risk assessment tool. Further studies on patients with chronic conditions may be helpful to validate this finding.

• 대한간호학회지
• /
• 제40권1호
• /
• pp.14-23
• /
• 2010

Purpose: This study was done to compare health risk behavior prevalence for youth living in metropolitan, medium sized and small cities or rural area, in order to enhance understanding regional differences. Methods: For this study, data from the 2006 Youth Health Risk Behavior Online Survey collected by the Korean Center for Disease Control were analyzed using SPSS. Results: In the metropolitan areas, prevalence for disease and perceived obesity were higher than in other areas. Lack of intense or moderate physical activity, obesity, fast food intake, and insufficient sleep showed higher prevalence than in rural areas. Prevalence of lifetime smoking, lifetime alcohol consumption, present alcohol use, fruit intake less than once a day, and not wearing a seat belt were higher in rural areas than in urban areas. Gender, smoking, and alcohol use were correlated. Spearman correlation between living with parent and skipping breakfast were significant. Smoking, alcohol use, and sexual behavior were correlated. Conclusion: As significant differences in prevalence of youth health risk behaviors exist between regional areas, health education and health promotion programs considering these differences have to be developed and implemented for adolescents. Programs for prevention of smoking and alcohol use, programs for improvement of fruit intake and safety are suggested for adolescents in rural areas, whereas programs to enhance physical activity and obesity management are suggested for adolescents in metropolitan areas.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권12호
• /
• pp.7271-7275
• /
• 2013

Objective: To investigate the association between the gene polymorphisms of angiotensin-converting enzyme (ACE) and digestive system cancer risk. Method: A search was performed in Pubmed, Medline, ISI Web of Science and Chinese Biomedical (CBM) databases, covering all studies until Sep 1st, 2013. Statistical analysis was performed by using Revman5.2 and STATA 12.0. Results: A total of 15 case-control studies comprising 2,390 digestive system cancer patients and 9,706 controls were identified. No significant association was found between the I/D polymorphism and digestive cancer risk (OR=0.93, 95%CI = (0.75, 1.16), P=0.53 for DD+DI vs. II). In the subgroup analysis by ethnicity and cancer type, no significant associations were found for the comparison of DD+DI vs. II. Results from other comparative genetic models also indicated a lack of associations between this polymorphism and digestive system cancer risks. Conclusions: This meta-analysis suggested that the ACE D/I polymorphism might not contribute to the risk of digestive system cancer.

• 기본간호학회지
• /
• 제19권4호
• /
• pp.444-452
• /
• 2012

Purpose: To explore the usefulness of previous fall history as a triage variable for inpatients. Methods: Medical records of 21,382 patients, admitted to medical units of one tertiary hospital, were analyzed retrospectively. Inpatient falls were identified from the hospital's self-report system. Non-falls in 1,125 patients were selected by a stratified matching sampling with 125 patients with falls (0.59%). A comparative and predictive accuracy analysis was conducted to describe differences between the two groups with and without a history of falls. Logistic regression was used to measure the effect size of the fall history. Results: The fall history group showed higher prevalence by 9 fold than the non-fall history group. The relationships between falls and relevant variables which were significant in the non-fall history group, were not significant for the fall history group. Falls in the fall history group were 25 times more likely than in the non-fall group. Predictive accuracy of the risk assessment tool showed almost zero specificity in the fall history group. Conclusion: The presence of fall history, the fall prevalence, variables relevant to falls, and the accuracy of the risk tool were different, which support the usefulness of the fall history as a triage variable.

• 건축역사연구
• /
• 제21권2호
• /
• pp.37-52
• /
• 2012

This research has set up the fire fighting general index for Fire fighting of Crowded Wooden Building Cultural Asset which is composed of traditional wooden building instinct or complex. The results of this study are as follows. First, Fire fighting general index for crowded wooden building cultural asset, it is necessary to set fire fighting priority by considering fire risk and cultural asset characteristic and establish the system to cope with fire disaster in the most effective way by arranging facilities with restricted resource. Second, Fire risk is the index to draw fire and spread risk of cultural asset by applying index calculation processes such as fire load, burning velocity and ignition material spread characteristic to various aspects such as individual building and complex and combining their results. Cultural asset importance index consists of individual building evaluation, publicity security degree, area importance evaluation and historical landscape degree evaluation. Third, for each index combination process, weight of each index is drawn on the basis of AHP analysis result that is performed to the specialists of related fields. The formula to apply and combine it is prepared to apply the model to include meaning of each index and comparative importance degree.