• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.67-75
• /
• 2018

Multidimensional scaling (MDS) is an exploratory analysis of multivariate data to represent the dissimilarity among objects in the geometric low-dimensional space. However, a general MDS map only shows the information of objects without any information about variables. In this study, we used MDS based on the algorithm of Torgerson (Theory and Methods of Scaling, Wiley, 1958) to visualize some clusters of objects in categorical data. For this, we convert given data into a multiple indicator matrix. Additionally, we added the information of levels for each categorical variable on the MDS map by applying the partition method of Shin et al. (Korean Journal of Applied Statistics, 28, 1171-1180, 2015). Therefore, we can find information on the similarity among objects as well as find associations among categorical variables using the proposed MDS map.

• Journal of Energy Engineering
• /
• v.15 no.1 s.45
• /
• pp.74-81
• /
• 2006

In this study, on the basis of an analytic hierarchy process (AHP) method and through a questionnaire on subjective preference and importance, various power supply systems were comprehensively compared with multiple decision criteria such as environmental, social, healthy, and economic viewpoints and then overall priority was assessed. When a decision-making problem is modelled by a hierarchy structure, the AHP method is regarded as a useful tool for extracting subjective opinions via the aforementioned questionnaire. Here, the overall preferences were obtained by linearly aggregating weighting vector and preference matrix. The energy systems such as nuclear, coal, and LNG power plants were selected because they took share over 90% of domestic electricity supply in Korea. Furthermore, wind power and photovoltaic solar systems were included as representative renewable energy systems in Korea. According to the results of this demonstration study, the following comprehensive comparison indicators were yielded: 1) weighting factors for 4 types of main criteria as well as for 11 types of sub-criteria; 2) preference valuation for 7 types of energy systems under consideration; 3) overall score for each energy systems.

• Journal of Preventive Medicine and Public Health
• /
• v.20 no.1 s.21
• /
• pp.67-83
• /
• 1987

This study relates to the problems of organizational performance in the Christian hospitals. In this study, quality of working life (QOWL), which harmonizes individual as well as organizational goals, was used as an indicator of organizational performance from the open systems view. In order to identify the behavioral factors influencing QOWL in hospitals, self-administered questionnaires were distributed to 1,926 employees who were randomly selected from fifteen Christian hospitals from August 1 to August 30, 1986. The following results were obtained: 1) All correlation coefficients between QOWL and behavioral variables were statistically significant even though their magnitude varied according to hospital size. 2) Using factor analysis, 32 variables were parsimoniously grouped into four factors: individual conflicts, group behavior, organizational characteristics and situation, and job characteristics. The proportion of variance explained by these factors ranged from 33.5% to 38.6% according to hospital size. 3) The overall effects of the four factors in the multiple logistic models ranged from 0.85 to 3.12 according to hospital size. Among three hospital models, the model for small hospitals showed the best statistical fit. 4) The most influential factor was organizational characteristics and situation with an odds ratio ranging from 1.99 to 3.02. Again, the odds ratio was the highest for small hospitals. 5) For large hospitals, the two main factor effects were statistically significant: organizational characteristics and situation, and job characteristics. For medium hospitals, all main factor effects except job characteristics were statistically significant. For small hospitals, all main factor effects except group behavior were statistically significant. However, a factor interaction effect was shown only for large hospitals where it was statistically significant. 6) To examine whether the four factors influence financial performance, the four factor scores from the two financial performance groups were compared using Mann-Whitney test. The test results showed that the organizational characteristics and situation factor score was significantly different only for small hospitals.

• Phonetics and Speech Sciences
• /
• v.4 no.1
• /
• pp.99-108
• /
• 2012

Recent research has identified that self-reported voice problems are a risk indicator for voice disorders. However, previous studies concerning the general population did not take into account the influence of gender on self-reported voice problems. The purpose of the present cross-sectional study was to determine the gender differences in risk factors of self-reported voice problems in the Korean adult population using national survey data. This study utilized data from the Korea National Health and Nutritional Examination Survey 2008. Subjects inclued 3,622 people (1,508 male and 2,114 female) aged 19 years and older living in the community. Data were analyzed using t-test, one-way ANOVA, and multiple logistic regression. The prevalence of self-reported voice problems was 5.9% in males, and 8.1% in females Females had higher incidents of self-reported voice problems than males. Adjusting for covariates, in males, age (OR=2.47, 95% CI: 1.07-5.70), pain and discomfort during the last two weeks (OR=3.64, 95% CI: 2.20-6.01) were independently associated with self-reported voice problems (p<0.05). In women, age (OR=1.96, 95% CI: 1.18-3.26), education (OR=2.09, 95% CI: 1.06-4.12), smoking (OR=2.70, 95% CI: 1.48-4.93), thyroid disorders (OR=2.58, 95% CI: 1.47-4.53), pain and discomfort during the last two weeks (OR=1.75, 95% CI: 1.21-2.54) were independently associated with self-reported voice problem (p<0.05). Self-reported voice problems related risk factors differed according to gender. These findings suggest that there needs to be different program strategies that reflect gender differences in self-reported voice problems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9A
• /
• pp.891-899
• /
• 2008

The cdma2000 1xEV-DO system controls the data rates of mobile terminals based on a binary overload indicator from the base station and a simple probabilistic model. However, this traffic control scheme has difficulty in controlling the reverse-link traffic load effectively and in guaranteeing a stable operation of the reverse link because each mobile terminal determines the next data rate autonomously. This paper proposes a new trafRc control scheme to improve the system stability, and analyzes the proposed scheme by modeling it as a discrete-time Markov process. The numerical results show that the maximum data rate of the proposed scheme is much higher than that of the conventional one. Moreover, the proposed scheme does not modify the standard physical channel structure, so it is compatible to the existing 1xEV-DO system.

• Korean Journal of Soil Science and Fertilizer
• /
• v.40 no.5
• /
• pp.429-434
• /
• 2007

This study was conducted to estimate site productivity of Quercus acutissima and Quercus mongolica by four forest climatic zones. We used site environmental variables (28 geographical and pedological factors) and site index as a site productivity indicator from nation-wide 23,315 stands. Based on multiple regression analysis between site index and major environmental variables, the best-fit multivaliate models were made by each species and forest climatic zone. Most of site index prediction models by species were regressed with seven to eight factors, including altitude, relief, soil depth, and soil moisture etc. For those models, three evaluation statistics such as mean difference, standard deviation of difference, and standard error of difference were applied to the test data set for the validation of the results. According to the evaluation statistics, it was found that the models by climatic zones and species fitted well to the test data set with relatively low bias and variation. Also having above middle of site index range, total area of productive sites for the two Quercus spp. estimated by those models would be about 6% of total forest area. Northern temperate forest zone and central temperate forest zone had more productive area than southern temperate forest zone and warm temperate forest zone. As a result, it was concluded that the regressive prediction with site environmental variables by climatic zones and species had enough estimation capability of forest site productivity.

• Korean Journal of Environment and Ecology
• /
• v.24 no.2
• /
• pp.178-185
• /
• 2010

The purpose of this study is to verify the effectiveness of the biotope model in applying and developing Korean urban forests. We found that there are 17 biotope assessment indicators, including forest layer structure, site conditions, ratio of broad-leaved trees, species richness, etc. In terms of correlation analysis between indicators, the stand ages and the period of space formation have the highest relativity(coefficient 0.684). On the other hand, indicators that have negative relativity are layer structure and risk, with a coefficient of -0.412. Ten models were developed for the multiple regression analysis. 10 variables(site conditions(X2), ratio of broad-leaved trees(X3) and so forth except layer structure(X1), species richness(X4)) were found to have a 95% significance level The results from comparing the regression model and adding-up estimation matrix, the most accurate one was Model 3, which has a 91.7% out of the 10 models. However more monitoring will be needed to improve the accuracy of models for the Korean urban forests in future.

• Journal of Environmental Impact Assessment
• /
• v.23 no.5
• /
• pp.379-392
• /
• 2014

This study was performed to predict the future climate envelope of Pinus pumila, a subalpine plant and a Climate-sensitive Biological Indicator Species (CBIS) of Korea. P. pumila is distributed at Mt. seorak in South Korea. Suitable habitat were predicted under two alternative RCPscenarios (IPCC AR5). The SDM used for future prediction was a Maxent model, and the total number of environmental variables for Maxent was 8. It was found that the distribution range of P. pumila in the South Korean was $38^{\circ}7^{\prime}8^{{\prime}{\prime}}N{\sim}38^{\circ}7^{\prime}14^{{\prime}{\prime}}N$ and $128^{\circ}28^{\prime}2^{{\prime}{\prime}}E{\sim}128^{\circ}27^{\prime}38^{{\prime}{\prime}}E$ and 1,586m~1,688m in altitude. The variables that contribute the most to define the climate envelope are altitude. Climate envelope simulation accuracy was evaluated using the ROC's AUC. The P. pumila model's 5-cv AUC was found to be 0.99966. which showed that model accuracy was very high. Under both the RCP4.5 and RCP8.5 scenarios, the climate envelope for P. pumila is predicted to decrease in South Korea. According to the results of the maxent model has been applied in the current climate, suitable habitat is $790.78km^2$. The suitable habitats, are distributed in the region of over 1,400m. Further, in comparison with the suitable habitat of applying RCP4.5 and RCP8.5 suitable habitat current, reduction of area RCP8.5 was greater than RCP4.5. Thus, climate change will affect the distribution of P. pumila. Therefore, governmental measures to conserve this species will be necessary. Additionally, for CBIS vulnerability analysis and studies using sampling techniques to monitor areas based on the outcomes of this study, future study designs should incorporate the use of climatic predictions derived from multiple GCMs, especially GCMs that were not the one used in this study. Furthermore, if environmental variables directly relevant to CBIS distribution other than climate variables, such as the Bioclim parameters, are ever identified, more accurate prediction than in this study will be possible.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.2
• /
• pp.178-186
• /
• 2015

Endurance time is very important indicator to estimate muscle fatigue. In the case of measuring endurance time directly, it is dangerous for subject to perform a test until the point of failure to main time force. Therefore, this paper presents the model to estimate endirance time using indirect measurements such as personal factors and anthropometrical data. Previous studies had shown that personal factors such as gender and age were not related to endurance time, but recently studies have shown that it is estimated by using independent variable or predictor such as GTA (Gravitational Torque of the horizontal, stretched arm) and MVC (Maximum Voluntary Contraction). The present study investigated variables to estimate endurance time using personal factors and anthrometrical data during isotonic contractions. Twenty five healthy subject volunteered for this study, and performed three test sessions of isotonic contraction exercises at 10~50% respectively. Afterward the correlation coefficient and p-values were compared among regression models using personal factors and anthropometrical data. The results demonstrated that multi-regression model had significant coefficient of correlation, and was useful estimate endurance time.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.27-35
• /
• 2023

In this paper, we propose a process of increasing productivity by applying a deep learning-based defect detection and classification system to the prepreg fiber manufacturing process, which is in high demand in the field of producing composite materials. In order to apply it to toe prepreg manufacturing equipment that requires a solution due to the occurrence of a large amount of defects in various conditions, the optimal environment was first established by selecting cameras and lights necessary for defect detection and classification model production. In addition, data necessary for the production of multiple classification models were collected and labeled according to normal and defective conditions. The multi-classification model is made based on CNN and applies pre-learning models such as VGGNet, MobileNet, ResNet, etc. to compare performance and identify improvement directions with accuracy and loss graphs. Data augmentation and dropout techniques were applied to identify and improve overfitting problems as major problems. In order to evaluate the performance of the model, a performance evaluation was conducted using the confusion matrix as a performance indicator, and the performance of more than 99% was confirmed. In addition, it checks the classification results for images acquired in real time by applying them to the actual process to check whether the discrimination values are accurately derived.