• Journal of Korea Planning Association
• /
• v.54 no.3
• /
• pp.106-118
• /
• 2019

In order to develop pedestrian navigation service that provides optimal pedestrian routes based on pedestrian satisfaction levels, it is required to develop a prediction model that can estimate a pedestrian's satisfaction level given a certain condition. Thus, the aim of the present study is to develop a pedestrian satisfaction prediction model based on three machine learning algorithms: Logistic Regression, Random Forest, and Artificial Neural Network models. The 2009, 2012, 2013, 2014, and 2015 Pedestrian Satisfaction Survey Data in Seoul, Korea are used to train and test the machine learning models. As a result, the Random Forest model shows the best prediction performance among the three (Accuracy: 0.798, Recall: 0.906, Precision: 0.842, F1 Score: 0.873, AUC: 0.795). The performance of Artificial Neural Network is the second (Accuracy: 0.773, Recall: 0.917, Precision: 0.811, F1 Score: 0.868, AUC: 0.738) and Logistic Regression model's performance follows the second (Accuracy: 0.764, Recall: 1.000, Precision: 0.764, F1 Score: 0.868, AUC: 0.575). The precision score of the Random Forest model implies that approximately 84.2% of pedestrians may be satisfied if they walk the areas, suggested by the Random Forest model.

• Korean Journal of Health Education and Promotion
• /
• v.19 no.1
• /
• pp.171-184
• /
• 2002

The purpose of this study was to investigate the relationship between self concept and depression of middle-aged women. The Subjects were 102 middle-aged women whose age range were 40-59 years old living in urban area. The instruments used in this study were self concept scale developed by Rosenberg and the depression scale developed by Zung. Data were collected done from July 1 to July 31, 2001 by a structured questionnaire. The Data were analyzed by the SPSS/PC+ program using t-test, ANOV A and Pearson Correlation Coefficient. The results of the study were as followed : 1. Mean score of self concept was 2.35. 2. Mean score of depression scale was 2.43 3. The relationship between self concept and depression was statistically negative correlation(r=-0.3769, P〈0.001). 4. There were statistically significant differences in the score of middle-aged women's self concept to the age(F=3.41, P〈0.05), education(F=2.97, P〈0.05), Occupation(t=1.84, P〈0.05). 5. General characteristics variables were significantly related to the level of depression were the age(F=3.12, P〈0.05), numbers of children(t=3.59,P〈0.05). Obstetrical characteristics variables were significantly related to the level of depression were age of menarche(F=4.03, P〈0.05), times of abortion(t=10.09, P〈0.001). In conclusion, this study revealed that self esteem was an important factor related to depression. I suggest that to decrease depression, it should be encouraged self esteem.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.10 no.3
• /
• pp.354-360
• /
• 2003

Purpose: This study was done for the purpose of identifying the relationship between work stress and job satisfaction of nurses who were evaluating medical insurance. Method: The participants in the study were 154 nurses who were in charge of evaluating medical insurance in 32 hospitals in Korea. Data were collected for 1 month from April 20 to May 20, 2001. Result: The mean total work stress score was 93.79. In comparison with work stress scores according to the work stress factors, 'work overload'(2.95) had the highest work stress score. The mean total job satisfaction score was 69.56. The total work stress scores did not show significant differences for any of the variables. The total job satisfaction score showed significant differences for the variables of age (F=3.51, p=.01), position (F=3.16, p=.02) and reason for transfer from one department to another (F=3.40, p=.003). Conclusion: The total work stress score showed a inverse correlation(v=-.36, p=.001) with the total job satisfaction score.

• Korean Journal of Adult Nursing
• /
• v.13 no.1
• /
• pp.82-95
• /
• 2001

The purpose of this study was to examine the relationship among perceived health status, self-esteem, self-efficacy and health promoting behavior, and to determine the predictors of health promoting behavior in post-mastectomy patients. The study, a descriptive correlational study, was done with structural questionnaires. A total of 51 post-mastectomy subjects from C university hospital in Kwang-ju, South Korea completed mail-in self-reporting questionnaires during a three month period from March to June, 1999. The data were collected using Lawstone's(1982) perceived health status scale, Rosenberg's(1965) self-esteem scale, the modified self-efficacy scale(Shere et al, 1982), and the modified health promoting lifestyle profile (Walker et al, 1987). The data obtained were analyzed according to percentage, mean and standard deviation, principal component analysis, varimax rotation, t-test, ANOVA, Pearson's correlation, and stepwise multiple regression. The results were as follows: 1. The health promoting behavior measurement resulted in six factors. Each factor was labelled as follows: self-actualization, nutrition, stress management, exercise, health responsibility and interpersonal support. The total percent of variance explained by the six factors was 58.4%. 2. The mean score of health promoting behavior was 85.92(range 58~117). The scores of six factor were nutrition 3.20, self-actualization 2.59, stress management 2.58, interpersonal support 2.58, health responsibility 2.49, and exercise 2.34 on a four point scale. 3. When the score of health promoting behavior factors were compared by general characteristics. Factor I: self-actualization, differed significantly by the frequency of pregnancy (F=3.06, p=.037). Factor II: nutrition differed significantly by drinking experience(t=-2.26, p=.028) and the pre- or post stage of menopause(F=2.69, p=078). FactorIII: stress management differed significantly depending on regularity of mensturation(t=-2.12, p= .042). FactorIV: exercise differed significantly by type of religion (F=2.49, p=.072), marital status(F=5.03, p=.010), and feeding type (F=2.64, p=.036). Factor V: health responsibility differed significantly by regularity of mensturation(t=2.18, p=.037). 4. The total health promoting behavior score was significantly related to self-esteem and perceived health status(r=.610, p.006; r= .378, p=.006). The score of selfactualization also corresponded with selfesteem and perceived health status(r=.556, p=.001; r=.343, p=.013). 5. The predictor to explain the score of health promoting behavior was self-esteem, which accounted for 37.1% of the total variance. The predictor to explain the score of self-actualization was self-esteem, which accounted for 30.9% of the total variance. The score of nutrition was primarily affected by both premenopause and drinking experience, which accounted for 13.1% and 9.5% respectively. Finally, the score of exercise was dictated by marriage, Buddhism, no experience of breast feeding, which accounted for 17%, 9.8%, & 5.2% respectively. In conclusion, self-esteem is the main predictor for health promoting behavior in post-mastectomy women. These findings suggest a need for nursing strategies which promote self-esteem in such patients.

• The Korean Journal of Rehabilitation Nursing
• /
• v.9 no.1
• /
• pp.15-24
• /
• 2006

Purpose: This study was to explore the cognitive function of Korean-Chinese stroke patient in China. Method: The study sample was 100 who were possible to communicate and agreed. The data were collected from one Brain's hospital at Yanji in China and by trained nurse from December 12, 2005 to April 28, 2006. The measurement tools were Digit span, Trail making, and MMSE-K. The data were analysed by SPSS Win 11.5 using frequency, t-test, ANOVA, and Pearson's correlation coefficients. Results: The mean score of DSF was 5.07, 3.42 of DSB, 161.37 of TMA, 229.28 of TMB, 22.64 of MMSE-K. There was a significant difference in DSF (F=6.35, p=.001), DSB (F=6.10, p=.001), TMA (F=3.53, p=.018), TMB (F=3.26, p=.025), MMSE-K score (F=12.97, p=.000) according to age, and DSF (F=6.67, p=.000), DSB (F=6.01, p=.000), TMA (F=5.82, p=.001), TMB (F=6.23, p=.001), and MMSE-K score (F=13.02, p=.000) according to educational level, and DSF (F=5.35, p=.006), DSB (F=3.16, p=.047), TMA (F=3.30, p=.041), TMB (F=3.42, p=.037), and MMSE-K score (F=4.95, p=.009) according to duration of disease, and DSB (F=3.54, p=.018), and MMSE-K (F=6.05, p=.001) according to frequencies of hospitalization. There was high correlation between DSF and DSB (r=.581, p=.000), TMA and TMB (r=.936, p=.000), MMSE-K and DSF (r=.579, p=.000), MMSE-K and DSB (r=.591, p=.000), DSF and TMA (r=.727, p=.000), and DSF and TMB (r=.721, p=.000). Conclusion: The cognitive evaluation score of Korean-Chinese stroke patients in China was in normal limit. The age, educational level, duration of disease and income were significant demographic characteristic affecting cognitive function. Further study need to compare the cognitive function of Korean-Chinese stoke patients in China and stoke patients in Korea.

• The Korean Journal of Rehabilitation Nursing
• /
• v.4 no.2
• /
• pp.219-231
• /
• 2001

The purpose of this study was to investigate the relationship among percieved social support, hope and quality of life of the cancer patients and to gain the baseline data for development of nursing intervention program for promoting quality of life in cancer patients. The design of this study was a cross sectional correlational survey. The subjects were 220 out and in-cancer patients in 5 general hospitals in Pusan. The data were collected from July 2 to August 1, 2001. The instruments were the Percieved social support scale(16 items, 5 point scale) had developed by Tae(1986), Hope scale(12 items, 4point scale) developed by Nowotny(1989) and Quality of life scale(31 items, 10 point scale) developed by Tae et al.(2000). The data was analyzed by the SPSS/PC+ program using frequency & percentage, item mean & standard deviation, t-test, ANOVA & Scheffe test, Pearson's correlation coefficient. The results of this study was as follows: 1) The item mean score of quality of life was $6.05{\pm}1.16$ (range 0-10). The highest score of subarea of the quality of life was the spiritual wellbeing area ($7.09{\pm}1.63$) and the lowest score was social wellbeing area ($5.53{\pm}1.65$). The mean score of perceived social support was $52.65{\pm}10.32$ (최저 1, 최고 80). The mean score of family support was $32.71{\pm}6.66$ (range 1-40) and the mean score of medical team support was $19.93{\pm}5.95$ (range 1-40). The mean score of Hope was $37.02{\pm}5.64$ (range 1-48). 2) There were statistically significant difference in the score of quality of life according to the life effect of religion(F=3.97, p=0.00), treatment method(F=2.94, p=0.01), area of diagnosis(F= 3.48, p=0.01), stage of disease (F=13.74, p=0.00). 3) There was significant correlation between perceived social support(r=0.44, p=0.000 ; family support ; r=0.334, p=0.000, medical support; r=0.395, p=0.000), hope(r=0.563, p=0.000) and quality of life. In conclusion, there was a significant relationship among perceived social support, hope and quality of life. Therefore perceived social support, hope intervention programs should be developed to improve the quality of life in cancer patients.

• Research in Community and Public Health Nursing
• /
• v.5 no.1
• /
• pp.64-80
• /
• 1994

The purpose of this study was to identify factors affecting family caregiver burden, and to identify the relationship between family caregiver burden and family function /self-care of psychiatric out-patient. These data were collected by questionnaire from September 20 to October 8, 1993. The subjects were 285 family caregiver of psychiatric out-patients. The instruments used in this study were Caregiver Burden Inventory(CBI) by Novak(1989), self-care performing by Yu(1992), and Family APGAR by Smilkstein(1979). The data were analyzed by cronbach's $\alpha$, mean, standard deviation, percentage, t-test, ANOVA, Pearson's correlation coefficient, and Stepwise Multiple Regression with SPSS /pc+ program. The result of this study were as follows ; 1. The means of family caregiver Burden revealed total 2.00, Time-Dependence Burden 78, developmental Burden 2.22, physical Burden 1.90, social Burden 1.43, emotional Burden 2.18, financial Burden 1.51. family caregiver burden score showed moderate level. time-dependence burden showed the highest score and social burden showed the lowest score. 2. The means of family function revealed total 5.67. 7 through high-21.4% (61), low through 3-38.6%(110). family function score showed moderate level. 3. The means of patient's self-care performance revealed total 137.71. self-care performance showed moderate level. 4. A ststistically significant correlation between family caregiver burden and patient's demographic variables, age (F=3.83, p<.01), marrital status(F=3.50, p<.01), job(F=3.17, p<.01), diagnosis(F=4.46, p<.01), income (F=4.46, p<.01). No significant differences between family caregiver burden and prevalent period, religion, sex (p>.05). S. A ststistically significant correlation between family caregiver burden and family's demographic variables, age (F=7.34, p<.01), sex(t=-2.63, p<.01), education level(F=7.61, p<.01), income (F=8.13, p<.01), relation with patient (F=6.92, p<.01), job(F=2.03, p<.05), medical service (F=3.89, p<.05), presence of chronically ill without patient(t=-2.01, p<.05) 6. Family function was the highest factor predicting family caregiver burden(R=.4168, $R^2=.1737$), low education level of family, patient's self-care, family income accounted for 36% in family caregiver burden.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.35-42
• /
• 2021

In this paper, we propose a deep learning-based person re-identification method using a three-dimensional RGB-Depth Xtion2 camera considering joint coordinates and dynamic features(velocity, acceleration). The main idea of the proposed identification methodology is to easily extract gait data such as joint coordinates, dynamic features with an RGB-D camera and automatically identify gait patterns through a self-designed one-dimensional convolutional neural network classifier(1D-ConvNet). The accuracy was measured based on the F1 Score, and the influence was measured by comparing the accuracy with the classifier model (JC) that did not consider dynamic characteristics. As a result, our proposed classifier model in the case of considering the dynamic characteristics(JCSpeed) showed about 8% higher F1-Score than JC.

• Journal of Korean Academy of Nursing
• /
• v.28 no.3
• /
• pp.773-785
• /
• 1998

This study aims at providing more qualitative care in family nursing practice. It is designed to analyze the degree of quality of life(QL) among families of the patients. The subjects consisted of 79 families of hospital hospice patients and 74 families of home-based hospice patients. The ages of the subjects were 17-74 years, at five university hospitals in Seoul, Inchon, and Kyung Gi Province, and one clinic in Chunchon. The data were collected from September, 1996 to August, 1997. The instrument used for the study was the Quality of Life Scale (QLS), which was composed of six factors, developed by Ro, You Ja. The analysis was done using t-test, ANOVA Scheff test, and Stepwise multiple regression. The results were as follows : 1. There were no statistically significant differences between the families of hospital hospice patients and the families of home-based hospice patients ; however, the mean score of the families of hospital hospice patients was higher than that of the families of home-based hospice patients. The scores on QLS ranged from 75 to 224 with a mean score of 140.58 in the families of hospital hospice patients. In the families of home-based hospice patients, the scores ranged from 79 to 214, showing a mean score of 135.25. Among six factors of QLS, family relationships showed the highest score in both groups, but economic life showed the lowest score in the families of hospital hospice patients, and emotional state showed the lowest score in those of home-based hospital patients. Self-esteem and relationship with the neighborhood were significantly higher in the families of hospital hospice patients than the families of home-based hospice patients (t=2.69, P= 0.008 ; 1=2.04, p=0.043). 2. In the families of hospital hospice patients, QL had significant relationship with family member's age(F=2.52, p =0.029), marital status (F=3.57, P=0.018), economic state(F=6.07, P=0.004), and education level(F=3.77, P=0.014). In the families of home-based hospice patients, QL had significant relationship with marital status (F=2.53, P=0.049), education level(F=4.35, P=0.007), occupation(F=3.93, P=0.002), and patient's age(F=2.73, P=0.020) 3. Economic status accounted for 17% of QL, and diagnosis accounted for an additional 7% of QL in the families of hospital hospice patients by means of stepwise multiple regression analysis. In the families of home-based hospice patients, relationships with patient accounted for 12% of QL. The findings showed that self-esteem and relationship with the neighborhood were significantly higher in the families of hospital hospice patients than the families of homed-based hospice patients and family relationships showed the highest value in QL. These findings should be considered in nursing practice.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.8 no.2
• /
• pp.189-198
• /
• 2001

The purpose of this study was done to identify the relationship between the level of pain and depression in patients with rheumatoid arthritis. The subjects for this study were 222 patients registered in H University Hospital Rheumatoid Arthritis Center, and the period of data collection was from July 20, 2000 to August 30, 2000. The research instruments used in this study were the Graphic Rating Scale of Pain and the CES-D for depression. The cronbach's ${\alpha}$ of the CBS-D scale was .89. Data analysis, was done by the SPSSWIN 10.0 program using descriptive statistics. The results are as follows. 1) The total pain score ranged from 0 to 147 with a mean score for pain in patients with rheumatoid arthritis of 72.64. 2) The total depression score ranged from 20 to 72 with a mean score of 39.86. 3) There was a significant difference in pain according to sex(F=5.26, p<.05) and education level (F = 3.59, p<.05). 4) There was a significant difference in depression scores according to sex (F=7 76, p<.05) and education level (F=3.02, p<.05). 5) The level of pain had a significant correlation with the level of education level(r=-.174, p<.01). The level of depression was significant correlation with the level of pain (r=.237. p<.01).