• Journal of the Korean Society of School Health
• /
• v.36 no.3
• /
• 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.

• Psychiatry investigation
• /
• v.15 no.11
• /
• pp.1037-1045
• /
• 2018

Objective The present study aimed to identify risk factors for future SI and to predict individual-level risk for future or persistent SI among college students. Methods Mental health check-up data collected over 3 years were retrospectively analyzed. Students were categorized as suicidal ideators and non-ideators at baseline. Logistic regression analyses were performed separately for each group, and the predicted probability for each student was calculated. Results Students likely to exhibit future SI had higher levels of mental health problems, including depression and anxiety, and significant risk factors for future SI included depression, current SI, social phobia, alcohol problems, being female, low self-esteem, and number of close relationships and concerns. Logistic regression models that included current suicide ideators revealed acceptable area under the curve (AUC) values (0.7-0.8) in both the receiver operating characteristic (ROC) and precision recall (PR) curves for predicting future SI. Predictive models with current suicide non-ideators revealed an acceptable level of AUCs only for ROC curves. Conclusion Several factors such as low self-esteem and a focus on short-term rather than long-term outcomes may enhance the prediction of future SI. Because a certain range of SI clearly necessitates clinical attention, further studies differentiating significant from other types of SI are necessary.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.127-135
• /
• 2023

Nowadays, the number of patients committing suicide due to depression and stress is rapidly increasing. In addition, if stress and depression last for a long time, they are dangerous factors that can cause heart disease, brain disease, and high blood pressure. However, no matter how modern medicine has developed, it is a very difficult situation for patients with depression and heart disease without special drugs or treatments. Therefore, in many countries around the world, studies are being actively conducted to determine patients at risk of depression and patients at risk of suicide at an early stage using electrocardiogram, oxygen saturation, and brain wave analysis functions. In this paper, in order to analyze these problems, a computer simulation was performed to determine heart disease risk patients by establishing heart disease hypothesis data. In particular, in order to improve the predictive rate of heart disease by more than 10%, a simulation using fuzzy inference was performed.

• Journal of Korean Academy of Nursing
• /
• v.39 no.5
• /
• pp.651-661
• /
• 2009

Purpose: This study was done using meta-analysis to examine 58 studies from studies published in the past eight years (2000 to 2007) that included variables related to adolescents' suicidal ideation. Methods: The materials for this study were based on 32 variables which were selected from masters' thesis, doctoral dissertation and articles from Journals of the Korean Academy of Nursing. Results: The classification consisted of 5 variables groups and 32 variables. In terms of effect size on risk, variables which were significant included psychological variables (0.668), socio-cultural variables (0.511), family environmental variables (0.405), school environmental variables (0.221), and personal characteristics variables (0.147). In terms of effect size on protection, variables which were significant included personal characteristics variables (-1.107), psychological variables (-0.526), family environmental variables (-0.264), and school environmental variables (-0.155). In terms of effect size on risk variables, psychological variables (0.668) were highest. In terms of effect size on protective variables, the variable of personal characteristic (-1.107) was the highest. Conclusion: While the results indicate possible risk and protective variables for suicidal ideation, but prediction is still difficult. Further study to compare adolescents with similar variables but no suicidal ideation and those with suicidal ideation is necessary.

• Journal of Korean Public Health Nursing
• /
• v.37 no.1
• /
• pp.125-135
• /
• 2023

Purpose: The number of patients suffering from depression is rapidly increasing worldwide, and by 2030, it is expected to pose a severe social and economic burden. Reports suggest that approximately 30% of subjects with symptoms of depression do not attempt treatment. Therefore, predicting the characteristics of subjects with depressive symptoms who have not even attempted counseling treatment is essential to increase the participation rate for such treatment. This study intends to predict the participation rates for psychological counseling treatment for depression among subjects with depressive symptoms. Methods: This study used data from the 2021 Korea Community Health Survey (KCHS). Data analysis was carried out using a decision tree to design a model that predicted participation in psychological counseling for depression. Results: The results showed that subjects aged 65 to 74 had difficulty understanding the explanations of medical staff even though they did not have cognitive impairment. Only 11.1% of this group received psychological counseling, which was the lowest rate among the various age groups. Among the subjects, 62.4% of those aged 19-44 or 45-64, who had suicidal thoughts and attempted suicide, received psychological counseling and this was the highest rate among the age groups surveyed. Conclusion: The identification of people showing depressive symptoms is crucial for encouraging them to undertake treatment. Also, proper depression-oriented medical services should be developed and implemented for people with depressive symptoms who exhibit a blind spot towards attempting treatment.

• Journal of Korea Artificial Intelligence Association
• /
• v.1 no.1
• /
• pp.1-6
• /
• 2023

This study developed models using decision forest, support vector machine, and logistic regression methods to predict and prevent suicidal ideation among Korean adolescents. The study sample consisted of 51,407 individuals after removing missing data from the raw data of the 18th (2022) Youth Health Behavior Survey conducted by the Korea Centers for Disease Control and Prevention. Analysis was performed using the MS Azure program with Two-Class Decision Forest, Two-Class Support Vector Machine, and Two-Class Logistic Regression. The results of the study showed that the decision forest model achieved an accuracy of 84.8% and an F1-score of 36.7%. The support vector machine model achieved an accuracy of 86.3% and an F1-score of 24.5%. The logistic regression model achieved an accuracy of 87.2% and an F1-score of 40.1%. Applying the logistic regression model with SMOTE to address data imbalance resulted in an accuracy of 81.7% and an F1-score of 57.7%. Although the accuracy slightly decreased, the recall, precision, and F1-score improved, demonstrating excellent performance. These findings have significant implications for the development of prediction models for suicidal ideation among Korean adolescents and can contribute to the prevention and improvement of youth suicide.

• The Korean Journal of Applied Statistics
• /
• v.27 no.2
• /
• pp.317-330
• /
• 2014

Bipolar disorder is a psychopathy characterized by manic and major depressive episodes. It is important to determine the degree of depression when treating patients with bipolar disorder because 810% of bipolar patients commit suicide during the periods in which they experience major depressive episodes. The Hamilton depression rating scale is most commonly used to estimate the degree of depression in a patient. This paper proposes using the Hamilton depression rating scale to estimate the effectiveness of patient treatment based on the linear mixed effects model and the transition model. Study subjects were recruited from the Seoul National University Bundang Hospital who scored 8 points or above in the Hamilton depression rating scale on their first medical examination. The linear mixed effects model and the transition model were fitted using the Hamilton depression rating scales measured at the baseline, six month, and twelve month follow-ups. Then, Hamilton depression rating scale at the twenty-four month follow-up was predicted using these models. The prediction models were then evaluated by comparing the observed and predicted Hamilton depression rating scales on the twenty-four month follow-up.

• The Journal of Information Systems
• /
• v.30 no.2
• /
• pp.29-56
• /
• 2021

Purpose Cyber deviance of adolescents has become a serious social problem. With a widespread use of smartphones, incidents of cyber deviance have increased in Korea and both quantitative and qualitative damages such as suicide and depression are increasing. Research has been conducted to understand diverse factors that explain adolescents' delinquency in cyber space. However, most previous studies have focused on a single theory or perspective. Therefore, this study aims to comprehensively analyze motivations of juvenile cyber deviance and to develop a predictive model for delinquent adolescents by integrating four different theories on cyber deviance. Design/methodology/approach By using data from Korean Children & Youth Panel Survey 2010, this study extracts 27 potential factors for cyber deivance based on four background theories including general strain, social learning, social bonding, and routine activity theories. Then this study employs econometric analysis to empirically assess the impact of potential factors and utilizes a machine learning approach to predict the likelihood of cyber deviance by adolescents. Findings This study found that general strain factors as well as social learning factors have positive effects on cyber deviance. Routine activity-related factors such as real-life delinquent behaviors and online activities also positively influence the likelihood of cyber diviance. On the other hand, social bonding factors such as community commitment and attachment to community lessen the likelihood of cyber deviance while social factors related to school activities are found to have positive impacts on cyber deviance. This study also found a predictive model using a deep learning algorithm indicates the highest prediction performance. This study contributes to the prevention of cyber deviance of teenagers in practice by understanding motivations for adolescents' delinquency and predicting potential cyber deviants.