• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• Safety and Health at Work
• /
• v.12 no.1
• /
• pp.66-73
• /
• 2021

Background: The objective was to assess the knowledge level, attitude, and practice of health care workers towards standard precautions, and to identify the related factors. Furthermore, it was attempted to identify the proportion of having the experience of needle stick injuries (NSIs) and associated factors among participants. Methods: A cross-sectional study was conducted in a teaching hospital among 233 health workers using a self-administrated questionnaire. The questionnaire included eight knowledge items, seven practice items, and five attitude items. Based on the mean score of each category, responses were grouped into "satisfactory" and "unsatisfactory". Univariate, bivariate, and multivariable logistic regression analyses were done. Results: The mean age of the participants 32.95 (SD ± 9.70) and 62.2% of them were women. 57.5% of the staff had a satisfactory level of correct knowledge (>5 correct answers), 37.3% had a satisfactory positive attitude (>3 correct answers), and 30.9% had a satisfactory practice (>3 correct answers) towards standard precautions. The occupation was one of the predictors as doctors were less likely to have satisfactory knowledge and practice compared to nurses (OR = 0.269, 95% CI: 0.10-0.70 and OR = 0.248, 95% CI: 0.08-0.77, respectively). Out of 174 participants, 31.6% of them reported experiencing NSIs and support staff were 71% less likely to experience NSIs compared to nurses & paramedics. Conclusion: The findings revealed a substandard adherence of standard precautions among participants, which highlighted the necessity of the provision of a periodic, tailored training program based on the occupation and risk exposure.

• International Journal of Computer Science & Network Security
• /
• v.21 no.1
• /
• pp.27-33
• /
• 2021

The whole world now is dealing with Coronavirus, and it has turned to be one of the most widespread and long-lived pandemics of our times. Reports reveal that the infectious disease has taken toll of the almost 80% of the world's population. Amidst a lot of research going on with regards to the prediction on growth and transmission through Symptomatic carriers of the virus, it can't be ignored that pre-symptomatic and asymptomatic carriers also play a crucial role in spreading the reach of the virus. Classification Algorithm has been widely used to classify different types of COVID-19 carriers ranging from simple feature-based classification to Convolutional Neural Networks (CNNs). This research paper aims to present a novel technique using a Random Forest Machine learning algorithm with hyper-parameter tuning to classify different types COVID-19-carriers such that these carriers can be accurately characterized and hence dealt timely to contain the spread of the virus. The main idea for selecting Random Forest is that it works on the powerful concept of "the wisdom of crowd" which produces ensemble prediction. The results are quite convincing and the model records an accuracy score of 99.72 %. The results have been compared with the same dataset being subjected to K-Nearest Neighbour, logistic regression, support vector machine (SVM), and Decision Tree algorithms where the accuracy score has been recorded as 78.58%, 70.11%, 70.385,99% respectively, thus establishing the concreteness and suitability of our approach.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.12-22
• /
• 2021

This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on k-fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, F₁-score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.23-28
• /
• 2022

This paper analyzed the influence of employment, human resource development, and labor practices on the corporate performance. Data were collected from the Korea Labor Institute's workplace panel survey(WPS) from 2017, and the analysis used 2,868 companies. This study employed operating profit as a corporate performance. Employment included open recruitment of new employees, evaluation of NCS job competency, and implementation of core human resources acquisition program. Human resource development consisted of incumbent training, job competency improvement evaluation, management program for low performer, emoloyee's career plan, and HRD using job analysis. Labor practices included guarantees for parental leave, guarantees for maternity leave, and support for childcare facilities. The analysis method used binominal logistic regression analysis for two groups of operating profit surplus and deficit companies. As a result of the analysis, it was possible to confirm the influence of employment, human resource development, and labor practices on performance. And the implications of employment, human resource development, and labor practices to improve corporate performance were discussed.

• Korean Journal of Childcare and Education
• /
• v.18 no.3
• /
• pp.19-36
• /
• 2022

Objective: This study aimed to analyze the changes and predictive factors of mothers' parenting competencies during their children's second to third grades in primary school. Methods: We used the data from the Panel study of Korean Parental Educational Involvement. We classified 373 mothers into three groups, 'reduced' parenting competency, 'maintained' parenting competency, and 'increased' parenting competency, and conducted one-way variance analysis and multinomial logistic regression analysis. Results: First, the mothers' parenting competency decreased between their children's 2nd year and 3rd year in primary school. Second, the 'reduced', 'maintained', and 'increased' groups differed from each other in the degree of change in self-system competency, level of understanding of school life, number of counseling sessions with homeroom teachers, and social networking. Third, the degree of change in self-system competency and social networking predicted the increase in mothers' parenting competency. The degree of change in self-system competency and the level of understanding of school life predicted the maintenance of mothers' parenting competency. Conclusion/Implications: This study, for the first time, has revealed the change in mothers' parenting competency and its predictive factors after the second year in primary school. How to support the growth of mothers' parenting competency was also discussed.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.318-331
• /
• 2022

The government has been promoting suicide prevention policies, but the elderly suicide rate has still not improved. This study focused on the role of local governments in solving suicide problems and analyzed three-year data from 2015 to 2017 at local governments level to investigate the relationship between suicide prevention policies and elderly suicide rates. Multiple regression analysis and logistic regression analysis was conducted to control social capital factors, demographic factors, and medical use factors that can affect the elderly suicide rate. As a result of the analysis, it was confirmed that suicide prevention ordinances were enacted and suicide prevention centers were established in areas with high suicide rates. In areas with high suicide rates, the suicide rate decreases if the elapsed period is long after the establishment of the center. From the perspective of suicide rates, it was analyzed that the local welfare support system was more affected. Accordingly, it was confirmed that the suicide prevention policy should be established in connection with the reinforcement of welfare policies

• Journal of Preventive Medicine and Public Health
• /
• v.55 no.1
• /
• pp.68-79
• /
• 2022

Objectives: This study investigated associations between ethnicity and malaria awareness in East Nusa Tenggara Province (ENTP), Indonesia. Methods: A community-based cross-sectional study was conducted upon 1503 adults recruited by multi-stage cluster random sampling. A malaria awareness questionnaire was used to collect data, according to which participants were classified as aware or unaware of malaria. Logistic regression was applied to quantify the strength of associations of factors with malaria awareness. Results: The participation rate in this study was high (99.5%). The participants were distributed relatively evenly among the Manggarai, Atoni, and Sumba ethnicities (33.0, 32.3, and 30.2%, respectively). Malaria awareness was significantly different amongst these groups; it was most common in the Manggarai ethnicity (65.1%; 95% confidence interval [CI], 59.9 to 70.3) and least common in the Sumba ethnicity (35.0%; 95% CI, 27.6 to 42.4). The most prominent factor influencing the malaria awareness in the Sumba and Manggarai ethnicities was education level, whilst it was socioeconomic status (SES) in the Atoni ethnicity. The likelihood of malaria awareness was significantly higher in adults with an education level of diploma or above (adjusted odds ratio [aOR], 21.4; 95% CI, 3.59 to 127.7 for Manggarai; aOR, 6.94; 95% CI, 1.81 to 26.6 for Sumba). Malaria awareness was significantly more common amongst high-SES adults in the Atoni group (aOR, 24.48; 95% CI, 8.79 to 68.21). Conclusions: Low education levels and low SES were prominent contributors to lower levels of malaria awareness in rural ENTP. Interventions should focus on improving malaria awareness to these groups to support the Indonesian government's national commitment to achieve a malaria elimination zone by 2030.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.711-722
• /
• 2021

The purpose of this study is to explore the effects of employment for people with developmental disabilities on their social activities. As a research method, cross-analysis, correlation analysis, and logistic regression analysis were performed using questionnaire data on employment status and social activity for 391 persons with developmental disabilities. As a result of the analysis, first, employment for people with developmental disabilities was found to be significant in social activities. In other words, 79.7% of those with developmental disabilities who were employed participated in social activities, and 39.3% of those who were not employed. Second, employment for people with developmental disabilities, whose personal characteristics were carefully verified as a control variable, had a significant effect on social activities and increased by 5.2 times. According to the results of the study, it can be seen that active employment support for people with developmental disabilities and policy efforts for employment are required for participation in social activities of people with developmental disabilities. Finally it was suggested to supported employment, expand standard workplaces for the disabled, introduce a job coach system, and establish a network with related organizations.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.291-298
• /
• 2022

This study is a descriptive study to investigate the association between current smoking, high-risk alcohol drinking and depressive symptoms among female college students. The participants were 515 female college students, and data collection was conducted through online and mobile surveys from September 2020 to August 2021. Multivariable ordinal logistic regression analysis was performed to investigate the association between current smoking, high-risk drinking and depressive symptoms, and as a result, current smoking was significantly associated with depressive symptoms (OR= 2.524, 95% CI=1.051-6.061). Therefore, in order to improve the depressive symptoms of female college students, adequate support such as reducing the smoking rate through active smoking cessation education and preparing various on-campus programs should be provided.