• Journal of Intelligence and Information Systems
• /
• v.7 no.2
• /
• pp.19-33
• /
• 2001

Link weight analysis approach is suggested as a heuristic for selection of input nodes in artificial neural network for bankruptcy prediction. That is to analyze each input node\\\\`s link weight-absolute value of link weight between an input node and a hidden node in a well-trained neural network model. Prediction accuracy of three methods in this approach, -weak-linked-neurons elimination method, strong-linked-neurons selection method and integrated link weight model-is compared with that of decision tree and multivariate discrimination analysis. In result, the methods suggested in this study show higher accuracy than decision tree and multivariate discrimination analysis. Especially an integrated model has much higher accuracy than any individual models.

• Journal of the Korean Society of Food Culture
• /
• v.20 no.2
• /
• pp.172-185
• /
• 2005

This study was carried out to investigate the information concerning dietary patterns and analysis of the various factors that influence obesity. The subjects of this study were 1,020 middle and high school students in Seoul. Subjects were classified into under weight, normal weight and over weight group by body mass index. We investigated eating habits, life habits, food behavior and food consumption. Data were collected by questionnair and analysed with the SAS program. The results of this study way are summarized and concluded as fellows; In the case of dietary pattern, over weight group showed significantly higher in skipping a meal than the other group. Also over weight group tend to eat fast. There were significant differences of food intake frequency score by body mass index. From the results of factor analysis of variable related to obesity, 4 factors were generated and the factors were named 'Food behavior related to obesity', 'Snack consumption pattern', 'Life habit', 'Family environment related to food habit'. These factors were associated with obesity. To maintain nutritional balance and health, we should implement to ensure good dietary patterns.

• Korean Journal of Community Nutrition
• /
• v.20 no.4
• /
• pp.281-290
• /
• 2015

Objectives: Obesity is known to influence physical and mental health as well as the general quality of life. The aim of this study was to evaluate the effect of obesity related quality of life on selecting a goal for weight management in overweight and obese female patients. Methods: A total of 140 overweight or obese (Body mass index $23kg/m^2$) female outpatients aged ${\geq}20$ and ${\leq}60$ years from one clinic participated in this study. Patients' desired weight (goal weight, ideal weight, satisfactory weight, acceptable weight and disappointed weight) and obesity related quality of life measures were evaluated. Univariate and multivariate analysis were performed to evaluate the effect of obesity related quality of life on goal weight reduction (%) and goal body mass index (BMI). Results: Mean BMI of overweight group, mild obesity group and severe obesity group were $62.0{\pm}4.8kg$, $68.5{\pm}5.5kg$ and $83.5{\pm}9.6kg$, respectively. Mean weight loss expectations of the three groups were $16.4{\pm}4.7%$, $19.5{\pm}5.3%$ and $30.2{\pm}6.8%$, respectively and goal weight was significantly different among the three groups. Severe obesity group had a lower total quality of life score including physical, work-related, daily living domains than overweight or mild obesity groups. In univariate and multivariate regression analysis, psychosocial domain of quality of life had an effect on goal weight reduction (%) and goal BMI. Conclusions: The results of this study demonstrated that the obese patients with poorer quality of life and psychosocial health tended to choose higher goal weight reduction and lower goal BMI.

• The Journal of Korean Obstetrics and Gynecology
• /
• v.34 no.3
• /
• pp.65-78
• /
• 2021

Objectives: This study was aimed to investigate the weight change of the high-risk group and the general maternal group and weight-relating factors in the early postpartum period. Methods: We retrospectively reviewed the medical records of those who received postpartum care from January 1, 2020, to December 31, 2020, in the postpartum care center affiliated with one Korean Medicine hospital. A total of 257 postpartum women's medical charts were included and divided into the high-risk group and the general maternal group. We investigated the weight changes and Body mass index (BMI) of the postpartum women and compared the difference between the two groups after taking the postpartum care. Finally, we used a Pearson correlation analysis to identify the weight-relating factors in the early postpartum period. Results: All the postpartum women showed the following results; 33.81±4.03 years old as the mean age; 22.23±3.28 as pre-pregnancy BMI; 58.21±9.18 kg of pre-pregnancy weight increased into 70.75±9.70 kg in the last month of pregnancy. Of the total 257 patients, 149 (58.0%) of high-risk pregnancy experience and 108 (42.0%) of general pregnancy were included. The edema index right after childbirth was significantly higher in the high-risk group than in the general maternal group (p<0.001), but there was no significant difference in BMI. After treatment with Korean medicine treatments, body weight, BMI, and edema index decreased significantly in both groups (p<0.01). As a result of correlation analysis, weight gain during pregnancy had a significant negative correlation with pre-pregnancy weight and pre-pregnancy BMI and a significant positive correlation with weight and BMI of the last month of pregnancy. In particular, pre-pregnancy BMI and body weight showed a significant negative correlation only in the high-risk group. Postpartum weight loss was significantly positively correlated with pre-pregnancy weight, pre-pregnancy BMI, weight & BMI of the pregnancy last month, weight gain during pregnancy, and decrease in edema (p<0.01) Conclusions: The weight during pregnancy of the high-risk group increased in inverse proportion to the pre-pregnancy BMI. The level of edema right after childbirth was significantly higher than that of the general maternal group, but showed a significant decrease after 2 weeks of Korean medicine treatments. Although it implicates the need for active Korean medicine treatments in the early postpartum period, further studies with controlled groups are needed.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.5 no.1
• /
• pp.135-139
• /
• 2002

By means of correlation analysis and path coefficient analysis, 18 characteristics of 325 conserved silkworm races were studied for the selecting efficiency on main silk quality characteristics. The result showed that selecting lines with heavier cocoon weight are beneficial to raise the weight of have. Choosing strains with higher cocoon shell percentage, longer filament length and better reelability percentage is good for increasing non-broken filament length. Selecting lines with medium daily weight of have produced in the fifth instar, medium cocoon shell percentage and simultaneously selecting heavier cocoon shell weight are favorable to promote reelability percentage. Choosing strains with heavier cocoon weight, higher cocoon shell percentage, heavier daily weight of bave produced in the fifth instar and heavier weight of bave but shorter filament length may hopefully breed out a race with thick filament size. And selecting strains with heavier weight of have and fine filament size are beneficial to increase filament length. Selecting lines with higher cocoon shell percentage and heavier weight of bave are favorable to raise raw silk percentage of cocoon. There are no significant direct or indirect correlation coefficient between viability and major characteristics of silk quality. It is, therefore, feasible to breed a silkworm race with excellent vitality and good silk quality as well.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.6
• /
• pp.153-172
• /
• 2018

Overloading of Commercial vehicles have been an important area of transportation as one of the main causes of pavement damage, bridge collapse, severe traffic accident, etc. In this study, we analyzed the effects of overweight prevention by analyzing overweight driving patterns and using weight sensors. First, we analyzed relevant literatures of overweight and surveyed the commercial weight sensors. Then we chose the typical type of overweight vehicles based of overweight enforcement data analysis. MEMs inclinometer weight sensor were installed to 10 test vehicles and data was collected by weight sensors and gps in real time. As a result of gross vehicle weight and axle weight analysis, it was found weight sensor could decrease overweight rate. However, since the number of samples of test vehicles is insufficient to represent the whole commercial vehicle, further studies are deemed possible through the extension test.

• Nutrition Research and Practice
• /
• v.17 no.2
• /
• pp.218-227
• /
• 2023

BACKGROUND/OBJECTIVES: Malnutrition during hospitalization is linked to increased morbidity and mortality, but there are insufficient studies observing clinical factors contributing to weight loss during hospitalization in Indonesia. This study was therefore undertaken to determine the rate of weight loss during hospitalization and the contributing factors. SUBJECTS/METHODS: This was a prospective study involving hospitalized adult patients aged 18-59 yrs, conducted between July and September 2019. Body weight measurement was taken at the time of admission and on the last day of hospitalization. The factors studied were malnutrition at admission (body mass index < 18.5 kg/m²), immobilization, depression (Beck Depression Inventory-II Indonesia), polypharmacy, inflammatory status (neutrophil-lymphocytes ratio; NLR), comorbidity status (Charlson Comorbidity Index; CCI), and length of stay. RESULTS: Totally, 55 patients were included in the final analysis, with a median age of 39 (18-59 yrs) yrs. Of these, 27% had malnutrition at admission, 31% had a CCI score > 2, and 26% had an NLR value of ≥ 9. In all, 62% presented with gastrointestinal symptoms, and depression was documented in one-third of the subjects at admission. Overall, we recorded a mean weight loss of 0.41 kg (P = 0.038) during hospitalization, with significant weight loss observed among patients hospitalized for 7 days or more (P = 0.009). The bivariate analysis revealed that inflammatory status (P = 0.016) was associated with in-hospital weight loss, while the multivariate analysis determined that the contributing factors were length of stay (P < 0.001) and depression (P = 0.019). CONCLUSIONS: We found that inflammatory status of the patient might influence the incidence of weight loss during hospitalization, while depression and length of stay were independent predictors of weight loss during hospitalization.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2020.06a
• /
• pp.157-158
• /
• 2020

In order to calculate construction costs properly, it is necessary to add the weight that reflects different worksite conditions. The implementation of the weight, however, is difficult because it is impossible to determine whether wight should be added in basic work or whether weight values are overlapped. Special worksite conditions further complicate the matter. Furthermore, overlapping implementation of weight values result in overestimation of construction costs. The current study clearly analyzed the weight value items in the current construction cost calculation standards, and analyzed the weight value items included under the basic productivity category to propose an improvement of weight standards. Basically, the estimating standards provide 140 weight value items, with different levels of weight given to each item. Among 1,333 items in the estimating standards, 140 include weight values. Some items have two types of weight values.

• The Korean Journal of Community Living Science
• /
• v.25 no.4
• /
• pp.473-486
• /
• 2014

This study examines the relationships between dietary self-efficacy, social support, depression, and weight-control behavior of children and adolescents. A total of 593 students (280 children and 313 adolescents) from schools in Seoul, Gyeonggi-do, Jeollabuk-do, and Daegu were surveyed using scales for dietary self-efficacy, social support, depression, and weight-control behavior. The data were analyzed through a frequency analysis, means and standard deviations, a t-test, a one-way ANOVA, the Tukey test, the Pearson's correlation, and a stepwise multiple regression analysis. According to the results, weight-control behavior was more likely in girls than boys, high-school students than elementary-school students, and obese/overweight students than normal/underweight students. The factors influencing children's weight-control behavior were general eating habits, the BMI, depression, food choice, and family support, in that order, and those influencing adolescents' weight-control behavior were general eating habits, gender, and the BMI, in that order. These results are expected to be useful as basic data for developing weight-control programs for children and adolescents.

• Journal of Korean Biological Nursing Science
• /
• v.14 no.2
• /
• pp.77-83
• /
• 2012

Purpose: The purpose of this study was to identify the self image according to body mass index in middle school students. Methods: The study included 260 students. Self image was measured using structured questionnaires. Statistical analysis was performed using t-test, ANOVA, Scheffe's test and multiple regression analysis. Results: The proportions of under weight, normal weight, and overweight were 36.1%, 53.9%, 10.0% respectively. Also, the mean of the self image score was 4.0. The score of self image was significantly higher in the normal weight group than the over weight group or the under weight group. This trend was also consistent in multiple regression. The score of self image was lower in high school grades, low perceived record, under-weight and over-weight. Conclusion: This study shows that many students are under weight. The score of self image was related with body image. So, level of obesity in each student should also be considered, to elevate the self image.