• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.5
• /
• pp.2539-2554
• /
• 2017

Regression-based image super resolution (SR) methods have shown great advantage in time consumption while maintaining similar or improved quality performance compared to other learning-based methods. In this paper, we propose a novel single image SR method based on hierarchical regression to further improve the quality performance. As an improvement to other regression-based methods, we introduce a hierarchical scheme into the process of learning multiple regressors. First, training samples are grouped into different clusters according to their geometry similarity, which generates the structure layer. Then in each cluster, a compact dictionary can be learned by Sparse Coding (SC) method and the training samples can be further grouped by dictionary atoms to form the detail layer. Last, a series of projection matrixes, which anchored to dictionary atoms, can be learned by linear regression. Experiment results show that hierarchical scheme can lead to regression that is more precise. Our method achieves superior high quality results compared with several state-of-the-art methods.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.18 no.4
• /
• pp.452-462
• /
• 2011

Purpose: The purpose of this study was to examine the factors affecting body image among undergraduate students. Method: The research design for this study was a descriptive survey design using a convenience sampling. Data collection was done using self-report questionnaires with 319 undergraduate students located in 3 cities, Seoul, Gangneung and Seosan. Pearson correlation coefficients and hierarchical multiple regression with the SPSS Win 12.0 Program were used to analyze the data. Results: In the hierarchical multiple regression analysis, gender, height, weight and college major were controlled. Body surveillance and body shame significantly predicted 72.3% of appearance orientation. Sociocultural attitudes toward appearance and self-esteem significantly predicted 33.5% of appearance evaluation. Self-esteem and body surveillance significantly predicted 15.9% of health orientation. Self-esteem significantly predicted 23.3% of health evaluation. Conclusion: Findings from this study provide a comprehensive understanding of body image and related factors in undergraduate students in Korea. However, further study with a larger random sample and more a detailed research design is necessary.

• Health Policy and Management
• /
• v.13 no.3
• /
• pp.129-147
• /
• 2003

This study is to grasp a trend of profitability classified by characteristics of hospitals and to analyze related factors. Subjects are 145 hospitals which have gotten the standardization audit by Korean Hospital Association during 1998-200l. Profitability was measured in the aspect of operation profit rate with operating margin to gross revenue as proxy variables. Independent variables were classified by general factors (ownership, number of beds, period of establishment, competition), financial factors (liabilities to total assets, current ratio, fixed ratio, total asset turnover, inventories turnover), and factors related to patient treatment (average length of stay, bed occupancy rate, new outpatient ratio, admission ratio of outpatients, number of patients per specialist, personnel costs per adjusted inpatient, administrative costs per adjusted inpatient). Hierarchical multiple regression analysis model was used in this study. As a result of hierarchical multiple regression analyzation of operating margin to gross revenue, adjustive $R^2$ of general factors was relatively more powerful. The factors had significant effect on operating margin to gross revenue were ownership(+), number of beds(+), competition(+), current ratio(+), fixed ratio(+), total asset turnover(+), personnel costs per adjusted inpatient(-).

• IE interfaces
• /
• v.23 no.1
• /
• pp.78-88
• /
• 2010

In this paper we develop a model for the performance evaluation of defense R&D test development projects based on analytic hierarchy process. First, evaluation indicators are collected through the related literature survey and a delphi inquiry method. Second, stepwise multiple linear regression is used for developing a hierarchical structure for analytic hierarchy process in the evaluation model, which can make the selected evaluation indicators of the hierarchical structure independent. Also we verify the effectiveness of proposed indicators of the performance evaluation by comparing with the existing evaluation indicators. The developed indicators for the performance evaluation is more reasonable and practical than the previous indicators on defense R&D test development projects.

• Journal of Agricultural Extension & Community Development
• /
• v.30 no.4
• /
• pp.187-197
• /
• 2023

The study aims to identify factors related to landscape structure and socioeconomic characteristics that influence adults' physical activity and health. The study analyzes data from 148,643 respondents from the Korea Community Health Survey conducted in 2021. The study measures various factors, including the frequency of physical activity, stress, depression, and landscape metrics of forest patches. Hierarchical multiple regression analysis was used, controlling for socio-demographic characteristics. Larger forest patches and more irregular shapes are associated with higher levels of physical activity among adults. The percentage and area of the forest patch are associated with fewer mental health complaints. Configuration-related landscape metrics such as shape or arrangement of the forest patch did not show a significant association with physical activity. Overall, the study provides insights into how specific characteristics of forest patches in urban areas may influence physical activity and mental health among adults. It emphasizes the importance of green spaces and their potential positive effects on both physical and mental well-being.

• Journal of Korean society of Dental Hygiene
• /
• v.12 no.5
• /
• pp.973-985
• /
• 2012

Objectives : The aim of this study is to investigate the possible impact factors on adults' self-reports of dry mouth and to develop strategies to improve oral health education policy. Methods : This study was conducted on a total of 622 self-administered questionnaires adult above 20 and under 65 years of age living in Seoul and Gyeonggi provinces. The final participants consisted of 443 adults without chronic illness, taking medications and wearing dentures. The Hierarchical Multiple Regression model with three stages was used to assess the association for exposure of interest, such as socio-demographics, health-related behavior, mental health and self-reported of dry mouth. Results : The participants reported mean score of dry mouth($6.32{\pm}4.47$), of which 191 were male($6.81{\pm}4.56$) and 252 were female($5.94{\pm}4.37$). Hierarchical Multiple Regression revealed that the score of dry mouth was shown to be significantly higher for the following people: Males, who were employed, unemployed, negative self-perceived general health, perceived stress, and participants who had no experience awareness of distress in two weeks. The explanatory power was 21.9%. The most powerful impact factor regarding to employment was shown to be negatively associated to dry mouth, and self-perceived general health, experiencing awareness of distressful in two weeks was also important factors. Conclusions : Based on these results in order to develop oral health education policy strategies for the prevention and management of dry mouth, there need to be considered for the employee.

• Journal of Digital Convergence
• /
• v.13 no.9
• /
• pp.479-487
• /
• 2015

The purpose of this study was to examine the factors affecting Obesity-Stress among women in their twenties. The research design for this study was a descriptive survey design using a convenience sampling. Data collection was done using self-report questionnaires with 216 women located in M city. Pearson correlation coefficients and Hierarchical multiple regression with the SPSS Win 18.0 Program were used to analyze the data. In the hierarchical multiple regression analysis, Perceived body shape and BMI significantly predicted. A result of examining the correlation between stress and obesity Stress and the variables, Appearance Orientation,(r=.386, p=<.001) Self-Esteem(r=.324, p=<.001) Body Evaluation(r=-453, p=<.001), Body Appearance Satisfaction(r=-.560, p=<.001). hierarchical multiple regression step 3 shows that variables explained 54.1%. As a result on the basis of, This will require a variety of training programs provided in order to reduce obesity stress.

• Journal of the Korean Applied Science and Technology
• /
• v.41 no.2
• /
• pp.349-362
• /
• 2024

This study examines the relationship between social-emotional competence, major satisfaction, and career preparation behaviors for nursing students and identifies factors influencing career preparation behaviors. Data collection was conducted from December 5, 2023, to January 3, 2024, targeting 197 nursing college students (2nd to 4th year) from universities located in S City, C Province, and K Province. The collected data were analyzed using SPSS 29.0 software, employing descriptive statistics, t-tests, ANOVA (post hoc analysis using Scheffe), Pearson's correlation, and hierarchical multiple regression. The results indicated statistically significant positive correlations among nursing students' social-emotional competence, major satisfaction, and career preparation behaviors. Hierarchical multiple regression analysis revealed that factors influencing career preparation behaviors included year of study, participation in career-related programs, social-emotional competence, and major satisfaction, with these variables explaining 40.3% of the variance in career preparation behaviors. Based on the results, it is necessary to develop and apply programs to strengthen social-emotional competencies considering the academic year of nursing student. And customized career programs should be established to increase major satisfaction.

• Journal of the Korean Society of School Health
• /
• v.31 no.1
• /
• pp.1-9
• /
• 2018

Purpose: This study was conducted to explore the relationship between aggression, self-esteem, and social withdrawal. Methods: This research is a cross-sectional survey. This study used the 5th-year data of the Korean Children & Youth Panel Survey (KCYPS) including 218 middle school students having delinquent behavior. The collected data was analyzed through $x^2$ test, t-test, ANOVA, Pearson's correlation, and hierarchical multiple regression, using SPSS/Win (ver. 23.0). Results: Significant positive correlations were found between aggression and social withdrawal and between aggression and self-esteem. Higher aggression was associated with higher social withdrawal and lower self-esteem. The results of the 2-step regression are as follows. Aggression was negatively correlated with self-esteem, whereas self-esteem was positively correlated with social withdrawals. The hierarchical multiple regression showed that 21% of the variance of aggression was significantly accounted for by self-esteem and social withdrawal. The most significant factor influencing aggression was social withdrawal. Conclusion: These results suggest that earlier screening and intervention programs to increase self-esteem and decrease social withdrawal for early adolescents should be developed to prevent aggression.

• Journal of Families and Better Life
• /
• v.21 no.2
• /
• pp.87-101
• /
• 2003

The purposes of this study were to analyze causes of wife abuse through married couples as the research subject. The study employed ecological perspective to the study of wife abuse as an multiple dimensional and integrated paradigm combined with isolated theories of other research. The subjects were 369 married couples who live in Gwangju area. Data were analyzed with Cronbach'α, factor analysis, basic statistics, paired-t test, 1-test, ANOVA, Duncan's test, correlation analysis, stepwise multiple regression analysis. and hierarchical regression analysis using the SPSS 10.0 for windows. The major findings were as follows; 1. The psychological abuse score, physical abuse score, and sexual abuse score were lower than median without exception. Hut prevalence rates of wife abuse were considerably serious; 91.9% psychological abuse, 44.4% physical abuse, and 53.7% sexual abuse. 2. The results of the hierarchical regression analysis indicated that the marital conflict had the strongest impact on wife abuse. And the attitude toward wife abuse of husband, hostility of husband, exposure experience of domestic violence during a growth period of husband, perception toward social violence of husband, exposure experience of domestic violence during a growth period of wife, drinking problem of husband, and interpersonal relationships stress of husband had influenced on wife abuse. These variables accounted for 49.5% of variance of wife abuse behaviors. As the result of the study, it concludes that the ecological systemic approach on the cause of wife abuse is useful as a theoretical instrument. Suggestions and implications are made for further research and practical application.