• Korean Journal of Acupuncture
• /
• v.33 no.3
• /
• pp.102-113
• /
• 2016

Objectives : This study is aimed at developing and discussing the prediction model of blood stasis pattern of traditional Korean medicine(TKM) using machine learning algorithms: multiple logistic regression and decision tree model. Methods : First, we reviewed the blood stasis(BS) questionnaires of Korean, Chinese, and Japanese version to make a integrated BS questionnaire of patient-reported outcomes. Through a human subject research, patients-reported BS symptoms data were acquired. Next, experts decisions of 5 Korean medicine doctor were also acquired, and supervised learning models were developed using multiple logistic regression and decision tree. Results : Integrated BS questionnaire with 24 items was developed. Multiple logistic regression models with accuracy of 0.92(male) and 0.95(female) validated by 10-folds cross-validation were constructed. By decision tree modeling methods, male model with 8 decision node and female model with 6 decision node were made. In the both models, symptoms of 'recent physical trauma', 'chest pain', 'numbness', and 'menstrual disorder(female only)' were considered as important factors. Conclusions : Because machine learning, especially supervised learning, can reveal and suggest important or essential factors among the very various symptoms making up a pattern identification, it can be a very useful tool in researching diagnostics of TKM. With a proper patient-reported outcomes or well-structured database, it can also be applied to a pre-screening solutions of healthcare system in Mibyoung stage.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.755-768
• /
• 2017

As Big Data becomes the core of the fourth industrial revolution, big data-based processing and analysis capabilities are expected to influence the company's future competitiveness. Comparative studies of RHadoop and RHIPE that integrate R and Hadoop environment, have not been discussed by many researchers although RHadoop and RHIPE have been discussed separately. In this paper, we constructed big data platforms such as RHadoop and RHIPE applicable to large scale data and implemented the machine learning algorithms such as multiple regression and logistic regression based on MapReduce framework. We conducted a study on performance and scalability with those implementations for various sample sizes of actual data and simulated data. The experiments demonstrated that our RHadoop and RHIPE can scale well and efficiently process large data sets on commodity hardware. We showed RHIPE is faster than RHadoop in almost all the data generally.

• Clinics in Shoulder and Elbow
• /
• v.16 no.1
• /
• pp.63-72
• /
• 2013

In medical research, multivariate analysis, especially multiple regression analysis, is used to analyze the influence of multiple variables on the result. Multiple regression analysis should include variables in the model and the problem of multi-collinearity as there are many variables as well as the basic assumption of regression analysis. The multiple regression model is expressed as the coefficient of determination, $R^2$ and the influence of independent variables on result as a regression coefficient, ${\beta}$. Multiple regression analysis can be divided into multiple linear regression analysis, multiple logistic regression analysis, and Cox regression analysis according to the type of dependent variables (continuous variable, categorical variable (binary logit), and state variable, respectively), and the influence of variables on the result is evaluated by regression coefficient${\beta}$, odds ratio, and hazard ratio, respectively. The knowledge of multivariate analysis enables clinicians to analyze the result accurately and to design the further research efficiently.

• The Journal of Korean Academy of Prosthodontics
• /
• v.46 no.6
• /
• pp.610-619
• /
• 2008

STATEMENT OF PROBLEM: Factors affecting patients' decision-making for dental prosthetic treatment should be examined in terms of understanding improving patients' oral health. PURPOSE: The main purpose of this dissertation was to investigate patients' dental prosthetic treatment and factors affecting patients' decision-making for dental prosthesis treatment in Deagu and Gyungbook areas. MATERIAL AND METHODS: This study was based on the preliminary survey of dental patients conducted from July 1 to August 31 in 2006. A total of 700 questionnaires had been distributed and 640 were collected. 629 questionnaires were used for the statistical analysis. Descriptive and inferential statistics, such as frequencies, cross tabulation analysis, correlation analysis, logistic regression analysis, and multiple regression analysis were introduced. In the multiple regression analysis and logistic regression analysis, twenty-two independent variables were employed to explore the factors which have impacts on decision-making and satisfaction. RESULTS: The results of this dissertation are as follows: Logistic regression analysis turned out that monthly income, age, degree of expectation, marital status, and employer-insured policy of national insurance statistically increased the odds of decision-making of dental prosthesis treatment. But educational attainment decreased the odds ratio of the decision-making of dental prosthesis treatment. However, the rest independent variables do not have statistically significant impacts on the decision-making of dental prosthesis treatment CONCLUSION: Among independent variables, marital status had the most significant influence on the decision making of dental prosthesis treatment. Finally, suggestions for the future study and policy implications to improve satisfaction of the patients' dental prosthetic treatment were discussed.

• Journal of Information Technology Services
• /
• v.23 no.2
• /
• pp.1-12
• /
• 2024

This study aims to find ways to increase the declining donation intention by using machine learning techniques. To this end, in order to predict factors that affect donations between the MZ generation and the senior generation, various machine learning algorithms, including logistic regression analysis, are applied to build a model to determine variables that affect donation intention, and provide statistical verification and evaluation indicators. In this study, differences in donation intention by generation were expected as a variable affecting donation intention, and the senior generation was expected to show a higher donation intention tendency than the younger generation. However, although the research results were not statistically significant, the younger generation showed a higher intention to donate, and these results are interpreted to mean that value consumption and ethical consumption, which are important to today's MZ generation, also influenced donations. However, there were differences between generations in the amount of donations, and higher donation amounts were confirmed among the senior generation (those in their 50s or older) than the younger generation. In addition, the results of the logistic regression analysis showed that previous donation experience had a positive effect on future donation intention, and the more motivation and importance of donation and various social participation activities online and offline, the more active one became in donating.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2004.04a
• /
• pp.597-602
• /
• 2004

This study examined the forecasting of instant messinger's users choice using neural network. We used the statistical methods which were Logistic Regression, MDA(Multiple Discriminant Analysis), and ANN(Artificial Neural Network). In the result, the forecasting performance of the ANN was better than conventional model(Logistic Regression, MDA).

• The Korean Journal of Emergency Medical Services
• /
• v.17 no.3
• /
• pp.169-179
• /
• 2013

Purpose: The performance rate to perform Cardiopulmonary Resuscitation (CPR) by witness in out-of-hospital Cardiac Arrest (OHCA) is very low in South Korea. To prevent the death caused by OHCA, it is important to encourage the witness to perform CPR actively. The purpose of the study is to investigate the influencing factors to affect bystander CPR rate. Methods: I conducted a questionnaire survey from 25 February to 4 March, 2013, receiving responses from 517 people in Korea. The questionnaire included social demographic factors, history of heart disease, knowledge of CPR, and the reliability of emergency medical service (EMS). A logistic regression analysis was conducted. Results: Among the 517 respondents, 294 (57.4%) had intention of performing CPR. Multiple logistic regression analysis found the following significant predictors of CPR intention: gender (odds ratio [OR] = 0.390), age (OR = 1.024), religion (OR = 0.843), and knowledge of CPR (OR = 4.734). Conclusion: This study indicated that the strongest predictor is knowledge of CPR. Therefore, it would be helpful to teach CPR nationwide to encourage performing CPR. In addition, effect of CPR education in religious facilities is necessary.

• Journal of Korean Academy of Nursing
• /
• v.36 no.5
• /
• pp.829-836
• /
• 2006

Purpose: The purpose of the present study was to identify prevalence of excessive daytime sleepiness(EDS) and its associations with sleep habits, sleep problems, depression, subjective health status and obesity in community dwelling adults. Method: Data was collected from adults aged 20-59 years by random sampling. Subjects completed a questionnaire which was composed of the Epworth Sleepiness Scale, Center for Epidemiologic Studies Depression Scale, and questions that included items about sleep habits, sleep problems, subjective health status and sociodemographic characteristics. Height and weight were measured for calculation of body mass index. The statistical analyses was based on 3,302 adults (51.5% males and 48.5% females). Descriptive statistics, univariate logistic regression and multiple logistic regression were used. Result: The prevalence of EDS was 17.1% Multiple logistic regression showed that the associated factors of EDS were depression, obesity, dissatisfaction with sleep time, irregular sleep, and habitual snoring. Depression was the most significant associated factor(adjusted odds ratio for severe depression=2.27, 95% Confidence Interval=1.73-2.96). Conclusion: EDS is a common symptom in adults. Our finding suggested that persons with a complaint of EDS should be completely assessed for depression and obesity as well as sleep problems.

• Journal of Korean Biological Nursing Science
• /
• v.25 no.4
• /
• pp.295-305
• /
• 2023

Purpose: This study compared health behaviors and health-related clinical characteristics between individuals with normal glucose levels without diabetes and those with undiagnosed diabetes. Factors that were associated with undiagnosed diabetes were identified by sex. Methods: This was an observational study with a cross-sectional design based on data from the eighth Korea National Health and Nutrition Examination Survey, which used a stratified, multi-stage, cluster-sampling design to obtain a nationally representative sample. Multiple logistic regression analysis was employed to compute the odds ratios of health behaviors and clinical characteristics to identify risk factors for undiagnosed diabetes. Results: The overall prevalence of undiagnosed diabetes was 5.2% (weighted %, n = 700, p < .001). Among individuals with undiagnosed diabetes, 58.3% were men. Univariate logistic regression for undiagnosed diabetes identified sex, age, house income, educational level, and triglycerides as influencing factors. In multiple logistic regression by sex, the factors associated with undiagnosed diabetes in men were age, perceived health status, a diagnosis of angina, and triglycerides. Conclusion: Strategies should be targeted to improve health behaviors and clinical characteristics for specific age groups, men in bad perceived health status, women with high systolic blood pressure, and high triglycerides. Moreover, healthcare providers should understand the barriers to health behaviors and health-related quality of life to effectively deliver healthcare services.

• Communications for Statistical Applications and Methods
• /
• v.26 no.5
• /
• pp.507-517
• /
• 2019

Statistical matching is a method of combining multiple sources of data that are extracted or surveyed from the same population. It can be used in situation when variables of interest are not jointly observed. It is a low-cost way to expect high-effects in terms of being able to create synthetic data using existing sources. In this paper, we propose the several statistical micro matching methods using a multinomial logistic regression model when all variables of interest are categorical or categorized ones, which is common in sample survey. Under conditional independence assumption (CIA), a mixed statistical matching method, which is useful when auxiliary information is not available, is proposed. We also propose a statistical matching method with auxiliary information that reduces the bias of the conventional matching methods suggested under CIA. Through a simulation study, proposed micro matching methods and conventional ones are compared. Simulation study shows that suggested matching methods outperform the existing ones especially when CIA does not hold.