Earticle

한국의 노인(65세 이상) 자살 문제는 점차 심각해지고 있는 추세이다. 급격한 인구 고령화 흐름에 따라 이러한 고령층의 자살 추세가 더욱 가속화될 것으로 추정되고 있어, 노인 자살을 예방하고 감소시키는 것이 개인 뿐만 아니라 중요한 사회적 과제로 대두되고 있다. 따라서 본 연구는 한국 노인들을 대상으로 자살 생각의 원인 요인을 파악하고 예측 모델을 개발하는 것을 목적 한다. 본 연구는 한국복지패널조사에서 제공하는 7개년의 패널 데이터를 활용하였으며 자살의 대인 관계 이론(interpersonal theory of suicide)과 사회 해체 이론(social disorganization theory)을 바탕으로 노인 자살의 잠재 원인 요인들을 선정한다. 다음으로 노인의 자살 생각에 대한 원인 요인 파악을 위해 패널 로짓 모형 분석을 진행하고 노인 자살 생각의 예측 모델 개발을 위해 딥 러닝과 머신 러닝 알고리즘을 활용한다. 본 연구는 계량 모형 분석을 통해 검증한 주요 원인 요인들을 활용하여 노인 자살을 예방할 수 있는 구체적인 노인 복지 정책 수립에 기여하고자 한다. 본 연구에서 제시된 예측 모델은 자살 고위험군 노인을 선별하고 관리할 수 있는 방안 마련의 기반을 제공한다. 또한 본 연구는 혼합방법론의 시너지를 보였다는 점에서 학술적 시사점을 가진다.

Elderly suicide problem has become worse in South Korea. With a rapid aging of the population, the trend of suicide among the elderly is expected to accelerate, preventing elderly suicide has been considered an important societal problem. Thus, we aim to investigate various factors that explain suicidal ideation and to develop a predictive model for suicidal ideation in the context of elderly people in South Korea. To this end, this study contributes to addressing the elderly suicide problem. By using seven-year panel data from the Korea Welfare Panel Survey, we extract various potential causal factors for elderly suicidal ideation based on interpersonal theory of suicide and social disorganization theory. Then a panel logit model was employed to assess the impacts of potential factors on suicidal ideation and deep learning and machine learning algorithms were used to develop a predictive model for suicidal ideation of elderly people. The results of our study provide practical implications for preventing elderly suicide by identifying causal factors of suicidal ideation and a high suicidal risk group of the elderly. This study sheds light on synergy of mixed methodology and provides various academic implications.

학생의 학업 성취를 사전에 예측하여 맞춤형 피드백과 지원을 제공하는 것은 학습 경험의 질을 향상시키고, 높은 성취를 유도하는 데 중요한 역할을 한다. 그러나 학업 성취도는 다양한 요인이 복합적으로 얽혀 있어 정확 한 예측이 어렵다. 기존 연구에서는 주로 종단연구 데이터를 활용하여 학업 성취에 영향을 미치는 변인을 탐색 해 왔으며, 최근에는 머신러닝 기법의 발전으로 다수의 변인과 비선형 관계를 동시에 분석함으로써 성취도 예측 의 정확성이 향상되고 있다. 특히, SHAP 지수를 활용한 연구를 통해 모델의 설명력을 높이고, 주요 변인의 교 육적 시사점을 시각화하는 시도가 이루어지고 있다. 본 연구는 대구교육종단연구 데이터를 활용하여 학생의 성 취도 변화를 시계열적으로 분석하고, 국어 및 수학 교과의 성취도 변화를 예측하는 주요 변인을 도출하여 학습 부진을 조기에 식별하는 데 기여하고자 한다.

Predicting students' academic achievement in advance and providing tailored support enhance learning quality and promote success. However, academic performance is influenced by complex factors, making accurate prediction challenging. While previous studies have used longitudinal data to explore key variables, recent advancements in machine learning improve accuracy by analyzing multiple factors and nonlinear relationships. In particular, SHAP-based studies enhance model interpretability and visualize key educational insights. This study uses data from the Daegu Education Longitudinal Study to analyze students’ academic performance over time and identify key predictors in Korean and mathematics, contributing to the early detection of underachievement.

Hospitals have accumulated large amounts of patient data, with each hospital’s data having unique characteristics and distributions. By leveraging this vast amount of data for machine learning, we can develop predictive models, such as those for predicting long-term outcomes in ischemic stroke patients and provide valuable information for treatment decisions. However, data privacy concerns prevent hospital data from being put together on a centralized server. This study investigates the applicability of federated learning for predicting long-term outcomes in ischemic stroke patients using data from Hallym University Sacred Heart Hospital in Pyeongchon and Hallym University Sacred Heart Hospital in Chuncheon. Patient outcomes are defined as favorable if the modified Rankin Scale (mRS) score is 0-2 and poor if the mRS score is 3-6. There are two tasks: one predicting patient outcomes at 3 months after stroke and the other predicting patient outcomes at 1 year after stroke. A simple deep neural networks model is used for implementation of the prediction model and the federated learning environment. In conclusion, the federated learning models using basic FedAVG and weighted averaging FedAVG achieved 99.4%-99.9% performance of traditional centralized learning models.

연구목적: 본 연구는 군에서 가장 많이 발생하는 교통사고의 예방을 위해 부대별로 교통사고가 발생할 확률을 사전에 예측하는 모형의 개발 방안을 제시하는 것이다. 연구방법: 이를 위해 CRISP-DM(Cross Industry Standard Process for Data Mining) 방법론을 적용하였다. CRISP-DM 프로세스는 6단계로 구성 되어 있고, 각 단계는 Waterfall Model처럼 일방향으로 구성되어 있지 않고 단계 간 피드백을 통하여 단 계별 완성도를 높이게 되어 있다. 연구결과: 전체 집단을 대상으로 기 구축된 사고조사 데이터와 동일한 데이터 세트(data set)를 구축하여 모델링한 결과 분류기준 0.5로 했을 때, 교통사고예측을 위한 모형의 정확도, 특이도, 민감도, AUC에서 의미있는 결과치를 도출하였다. 결론: 예측모형을 설계하는 과정에 서 데이터의 부족으로 인해 의미 있는 예측값을 얻기 어려운 문제점이 확인되었다. 이를 해결하기 위해 합리적 추론이 가능한 데이터 세트(data set)를 재구성 및 확대하여 데이터 부족을 해소하고, 이를 활용 한 예측모형을 설계할 수 있는 방법론을 제시하였다.

Purpose: This study proposes a method for developing a model that predicts the probability of traffic accidents in advance to prevent the most frequent traffic accidents in the military. Method: For this purpose, CRISP-DM (Cross Industry Standard Process for Data Mining) was applied in this study. The CRISP-DM process consists of 6 stages, and each stage is not unidirectional like the Waterfall Model, but improves the level of completeness through feedback between stages. Results: As a result of modeling the same data set as the previously constructed accident investigation data for the entire group, when the classification criterion was 0.5, Significant results were derived from the accuracy, specificity, sensitivity, and AUC of the model for predicting traffic accidents. Conclusion: In the process of designing the prediction model, it was confirmed that it was difficult to obtain a meaningful prediction value due to the lack of data. The methodology for designing a predictive model using the data set was proposed by reorganizing and expanding a data set capable of rational inference to solve the data shortage.

구글의 인플루엔자 의사환자(ILI) 예측 서비스 시작 이래로 웹 검색 정보를 활용한 ILI 예측 연구들이 급속도로 확산되고 있는 가운데, 본 연구는 ILI 자료와 웹 검색 정보를 활용한 한국 ILI 단기 예측 모형을 개발해 성능을 평가해 보고자 한다. 한국에 특화된 ILI 예측 모형 개발을 위해 한국질병관리본부의 ILI 감시 자료와 구글 및 네이버의 한국어 검색 정보를 ARIMA 모형과 함께 사용하였다. 모형1은 ILI 자료만 사용하였으며, 모형 2와 3은 모형1에 구글과 네이버의 검색자료를 각각 추가하였다. 모형4는 모형 2와 3의 공통 검색어를 모형1에 추가하였다. 모형 훈련기간 동안 모든 예측모형들이 95%(R2) 이상의 높은 적합도를 보였으며, 예측기간 1과 2에서 모형1이 가장 우수한 예측력(99.98%, 96.94%)을 보였다. 모형 3(a)와 4(b, c)는 전체 예측기간에서 90% 이상의 안정적인 예측력을 보였지만, 모형1의 성능에는 미치지 못하였다. 본 연구에서 정확하고 안정적인 예측력을 보인 모형들은 성능개선에 관한 보완적 연구와 더불어 국내 인플루엔자 유행 조기경보 시스템에 활용 가능하다.

Since Google launched a prediction service for influenza-like illness(ILI), studies on ILI prediction based on web search data have proliferated worldwide. In this regard, this study aims to build short-term predictive models for ILI in Korea using ILI and web search data and measure the performance of the said models. In these proposed ILI predictive models specific to Korea, ILI surveillance data of Korea CDC and Korean web search data of Google and Naver were used along with the ARIMA model. Model 1 used only ILI data. Models 2 and 3 added Google and Naver search data to the data of Model 1, respectively. Model 4 included a common query used in Models 2 and 3 in addition to the data used in Model 1. In the training period, the goodness of fit of all predictive models was higher than 95% (R2). In predictive periods 1 and 2, Model 1 yielded the best predictions (99.98% and 96.94%, respectively). Models 3(a), 4(b), and 4(c) achieved stable predictability higher than 90% in all predictive periods, but their performances were not better than that of Model 1. The proposed models that yielded accurate and stable predictions can be applied to early warning systems for the influenza pandemic in Korea, with supplementary studies on improving their performance.

Recently, the exhibition industry has developed rapidly with the development of information technologies. Most exhibitors in an exhibition plan and deploy many events that may provide advantages to visitors as a method of effective promotion. The growth and propagation of wireless technologies is a powerful marketing tool for exhibitors. However, exhibitors still rely on domain experts who are costly and time consuming because of the manual knowledge input procedure. Moreover, it is prone to biases and errors and not suitable for managing fast-growing and tremendous amounts of data that far exceed a human's ability to comprehend. To overcome these problems, data mining technology may be a great alternative, but it needs to be fit to each exhibition. This study uses data mining technology with the Predictive Model Markup Language (PMML) to suggest a system that supports intelligent services and that improves stakeholder satisfaction. This system provides advantages to the exhibitor, show organizer, and system designer, and is first enhanced by integrating data mining technologies through the knowledge of exhibition experts. Second, using the PMML, the system can automate the process of applying data mining models to solve real-time processing problems in the exhibition environment.

Purpose: Person-centered care emphasizes the therapeutic relationship between medical staff and patients, founded on mutual trust and understanding. In intensive care settings, there is growing recognition of the need to improve the care environment and promote patient-focused nursing. This study aimed to construct and validate a predictive model explaining person-centered care in intensive care units. Methods: This study employed a cross-sectional design involving 230 intensive care unit nurses working in a tertiary hospital, each with more than one year of direct patient care experience. Data were collected online between March 2 and March 30, 2023. Data analysis was conducted using IBM SPSS ver. 26.0 and AMOS ver. 25.0. Results: Statistically significant pathways were identified from nursing competency to the nursing work environment and person-centered care; from communication competence to teamwork and person-centered care; from nursing professionalism to teamwork and the nursing work environment; and from the nursing work environment to person-centered care. Nursing professionalism indirectly influenced person-centered care through teamwork and the nursing work environment. Conclusion: Enhancing person-centered care in intensive care units requires recognizing the critical roles of communication competence, nursing competency, and the nursing work environment. Developing and implementing educational programs that strengthen communication and nursing competencies, alongside initiatives that improve the nursing work environment, are essential.

Purpose: This study was conducted to assess predictor variables of husbands in multicultural families and examine the relationship among variables after setting up a hypothetical model including influencing factors, so as to provide a framework necessary for developing nursing interventions of domestic violence. Methods: The participants were 260 husbands in multicultural families in four cities in Korea. Data were analyzed using SPSS 22.0 and AMOS 20.0. Results: Self-control, social support, family of origin violence experience and stress on cultural adaptation directly affected to dysfunctional communication, and the explanatory power of the variables was 64.7%. Family of origin violence experience in domestic stress on cultural adaptation, and dysfunctional communication were directly related to domestic violence in multicultural families, and the explanatory power of the variables was 64.6%. We found out that all variables in the model had mediation effects to domestic violence through dysfunctional communication. In other words, self-control and social support had complete mediation effects, and family of origin violence experience in domestic violence and stress on cultural adaptation had partial mediation effects. Conclusions: The variables explained in this study should be considered as predictive factors of domestic violence in multicultural families, and used to provide preventive nursing intervention. Our resutls can be taken into account for developing and implementing programs on alleviating dysfunctional communication in multicultural families in Korea.

Purpose: This study investigated prediction of North Korean refugees' adaptation to the South Korean society and verified the goodness of fit between a hypothetical model and actual data in order to suggest the best model. Methods: This survey was conducted with 445 North Korean refugees living in a metropolitan area. Data were collected from September 1st to November 20th, 2012, and analyzed using SPSS Windows 18.0 and AMOS 17.0. Results: Traumatic experiences of North Korean refugees increased self-efficacy and psychological trauma. Acculturation stress decreased self-efficacy and increased passive coping. Self-efficacy affected active and passive coping, decreased psychological trauma, and increased resilience. Resilience is successful adaptation and refers to North Korean refugees' abilities to adapt effectively to stress. In particular, self-efficacy as the main parameter affecting resilience was confirmed. Conclusions: The results suggest that resilience can be improved through self-efficacy. It was the most significant factor decreasing psychological trauma and increasing resilience. Therefore, we need to develop programs for self-efficacy. The results also provide basic data for policy making for North Korean refugees.

Purpose: This study was conducted to develop and test a hypothetical model to predict health outcomes in young people with type 2 diabetes. Methods: Data were collected from 190 adults aged 23e45 with type 2 diabetes mellitus who visited the endocrinology outpatient department of the two university hospitals in South Korea from November 2, 2012 to March 7, 2013. Data collection used the structured questionnaires and patient medical records. The descriptive and correlation statistics were analyzed using PASW 18.0 and structural equation modeling procedure was performed using the AMOS 18.0 program. Results: The fit of the hypothetical model was appropriate with the ratio of the chi-square statistic to degrees of freedom at 17.00, goodness-of-fit index at .975, adjusted goodness-of-fit index at .930, root mean square error of approximation at .061, normed fit index at .926, Turker-Lewis index at .929, comparative fit index at .966. Behavioral skills were a critical factor that directly affects self-management behaviors. Through behavioral skills, motivation had a statistically significant indirect effect on selfmanagement behavior. Self-management behavior had a statistically significant direct effect on health outcome. Through self-management behavior, behavioral skills had a statistically significant indirect effect on health outcome. These variables explained 17.9% of the total variance for the health outcome in young people with type 2 diabetes. Conclusions: The results suggest that self-management behavior could be improved through nursing interventions promoting personal motivation (positive attitude), social motivation (social support), and behavioral skills (self efficacy), which can result in better health outcomes for young people with type 2 diabetes.

Purpose: This study aimed to develop a machine learning model to predict exclusive breastfeeding during the first 3 months after birth and to explore factors affecting breastfeeding outcomes. Methods: Data from 2,579 participants in the Korean Early Childhood Education & Care Panel between March 1 and June 3, 2025 were analyzed using Python version 3.12.8 and Colab. The dataset was split into training and testing sets at an 80:20 ratio, and five classifiers (random forest, logistic regression, decision tree, AdaBoost, and XGBoost) were trained and evaluated using multiple performance metrics and feature importance analysis. Results: The confusion matrix of the random forest classifier model demonstrated strong performance, with a precision of 86.6%, accuracy of 84.8%, recall of 96.8%, F1-score of 91.9%, and an area under the curve of 86.0%. Twenty-one features were analyzed, from which feeding plan, breastfeeding at 1 month, marriage period, maternal prenatal weight, self-respect, alcohol consumption, grit, value placed on children, maternal age, and depression emerged as important predictors of exclusive breastfeeding in the first 3 months. Discussion: A robust model was developed to predict exclusive breastfeeding that identified feeding planning and breastfeeding at 1 month as the most influential predictors. The model could be implemented in clinical and community settings to guide tailored breastfeeding support strategies, coupled with the integration of maternal self-respect, grit, and the value placed on children in counseling programs to promote exclusive breastfeeding.

Cement stabilisation is a common method for stabilising recycled road base material and provides a longer pavement life. With cement effect, the increment of stiffness in the stabilised layer would provide better load transfer to the pavement foundation. The recycling method provides an environmentally option as the existing road base materials will not be removed. This paper presents a case study of a trial section along the North-South Expressway in West Malaysia, where the Falling Weight Deflectometer (FWD) was implemented to evaluate the compressive strength and in-situ stiffness of the cement stabilised road base material. The improvement in stiffness of the cement stabilised base layer was monitored, and samples were tested during the trial. FWD was found to be useful for the structural assessment of the cement-stabilised base layer prior to placement of asphalt layers. Results from the FWD were applied to verify the assumed design parameters for the pavement. Using the FWD, an empirical correlation between the deflection and the stiffness modulus of the pavement foundation is proposed.

This paper presents methods of model predictive control (MPC) for eco-driving minimizing trip-time and energy- consumption of electric vehicles (EVs). Considering both non- convex and convex optimization problem formulations for MPC- based eco-driving, we compare the performances of nonlinear and linear MPC solutions for high-level planning of vehicle speed and charging in a driving simulation of a Munich– Cologne trip (573 km). The linear MPC can be considered as a convex quadratic programming approximation (i.e., convexified quadratic program) of the original nonlinear MPC, but its performance of optimality is shown to be comparable to the nonlinear counterparts whereas its computation speed is one order of magnitude faster.

This paper presents a model predictive control (MPC) based on a neural network (NN) model for air/fuel ration (AFR) control of automotive engines. The novelty of the paper is that the severe nonlinearity of the engine dynamics are modelled by a NN to a high precision, and adaptation of the NN model can cope with system uncertainty and time varying effects. A single dimensional optimization algorithm is used in the paper to speed up the optimization so that it can be implemented to the engine fast dynamics. Simulations on a widely used mean value engine model (MVEM) demonstrate effectiveness of the developed method.

본 연구는 WHO의 ICF(International Classification of Functioning, Disability and Health) 모델을 기반으로 재가노인의 도구적 일상생활수행능력 모형을 구축하였다. 연구 대상자는 65세 이상의 재가노인 260명이었고, 자료 분석은 SPSS Windows 18.0과 AMOS 18.0을 이용하였다. 모형의 적합도 검정 결과 모든 지수 기준을 충족하여 모형을 수용하기에 적합한 것으로 판단하였다. 연구 결과 재가노인의 도구적 일상생활수행능력에 직접적인 영향력이 가장 큰 요인은 신체 심리적 상태였다. 즉, 주관적 건강상태가 좋을수록, 우울이 낮을수록, 만성질환이 적을수록 도구적 일상생활수행능력이 좋았다. 직접적인 영향력이 두 번째로 큰 요인은 시각-운동 통합능력으로 확인되었고, 세 번째로 큰 요인은 사회활동이었다. 도구적 일상생활수행능력에 간접 영향을 미치는 변수는 개인 요인, 사회적 지지, 사회활동 순이었다. 도구적 일상생활수행능력에 대한 이들 변수의 설명력은 32%였다.

This study was conducted to construct and test a structural equation model of instrumental activities of daily living(IADL) in community-dwelling elderly. The model was based on ICF(International Classification of Functioning, Disability and Health) model. The participants were 260 elderly who were more than 65 years old. Physical and psychological function, visual-motor integration and social activities had direct effects on IADL. That is, the better the subjective health status, the lower the depression and the less chronic illness, the better IADL. Personal factor, social support and social activities had indirect effect on IADL. This model explained 32% of the variance in IADL.

자궁경부암은 전 세계적으로 여성에게 발생하는 암 중 네 번째로 흔한 암이며, 2020년 한 해 동안 60만 4천 건 이상의 신규 케이스가 보고되었고 이로 인한 사망자 수는 약 34만 1천 831명에 달했다. 콕스 회귀 모델은 암 연구에서 널리 채택되고 있는 주요 모델이지만, 비선형 연관성의 존재를 고려하면 선형 가정으로 인해 한계에 부딪힌다. 이러한 문제를 해결하기 위해, 본 논문에서는 ResNet의 잔여 학습 프레임워크를 활용하여 자궁경부암 사망률 예측의 정확성을 개선한 새로운 모델인 ResSurvNet을 제안한다. 이 모델은 본 연구에서 비교한 DNN, CPH, CoxLasso, Cox Gradient Boost, RSF 모델들을 능가하는 정확도를 보여주었기에 이러한 우수한 예측 성능은 자궁경부암 환자 관리에 있어 조기 진단 및 치료 전략 수립에 기여할 수 있고 임상적으로 적용할 때 큰 가치가 있음을 입증하며, 생존 분석 분야에서도 의미 있는 진전을 나타낸다.

Cervical cancer is the fourth most common cancer in women worldwide, and more than 604,000 new cases were reported in 2020 alone, resulting in approximately 341,831 deaths. The Cox regression model is a major model widely adopted in cancer research, but considering the existence of nonlinear associations, it faces limitations due to linear assumptions. To address this problem, this paper proposes ResSurvNet, a new model that improves the accuracy of cervical cancer mortality prediction using ResNet's residual learning framework. This model showed accuracy that outperforms the DNN, CPH, CoxLasso, Cox Gradient Boost, and RSF models compared in this study. As this model showed accuracy that outperformed the DNN, CPH, CoxLasso, Cox Gradient Boost, and RSF models compared in this study, this excellent predictive performance demonstrates great value in early diagnosis and treatment strategy establishment in the management of cervical cancer patients and represents significant progress in the field of survival analysis.

Purpose: This study aimed to identify the key factors influencing medication adherence among patients with hypertension and diabetes using data from the 2021 Korea Health Panel Survey (KHP) and to develop a predictive model for intervention planning. Methods: Secondary analysis was conducted using data from 3,376 hypertension and 1,497 diabetes patients. SPSS and SPSS Modeler were employed for descriptive statistics and decision tree analysis with the Chi-squared Automatic Interaction Detector (CHAID) algorithm to determine the major predictors of medication adherence. Results: For hypertension patients, adherence was influenced by health literacy, outpatient visits, adverse drug reactions, number of chronic diseases, disability, and the use of mental health counseling. For diabetes patients, adherence was affected by regular healthcare provider status, perceived stress, depressive symptoms, self-rated health, and unmet medical needs. Conclusion: A decision tree model revealed that higher health literacy, frequent healthcare use, and absence of adverse drug effects improved adherence. Psychological factors such as stress and depression were also significant. These findings emphasize the need for targeted programs that enhance health literacy and address emotional factors to improve adherence among patients with chronic diseases.

Purpose: The purpose of the study was to develop a predictive model of Health-related Quality of Life (HRQoL)for Korean Goose daddies - they live alone in Korea to support their families who moved overseas for children’s education. Methods: Data were collected from 151 goose daddies from May to June of 2011 by using the structured self-reported questionnaires. The collected data were analyzed using SAS program (version 9.2) and SAS CALIS procedure. Results: Frequency of exercise, monthly income, depression, perceived physical health, and perceived mental health had direct effects on HRQoL and Depression was the variable accounting for major total effect on HRQoL. It could be explained that predictor variables accounted for 76% of the health-related quality of life. Conclusion: In order to improve Goose daddies’ HRQoL, predictive factors, such as age, exercise, nutritional status, monthly income, depression, perceived physical health, and perceived mental health, should be considered. Furthermore, should the need of the exercise and diet program, early detection of depression and the treatment for it be emphasized. Also, there is a need to establish institutional structures to support goose daddies in adversity.