• Journal of Preventive Medicine and Public Health
• /
• v.38 no.2
• /
• pp.132-140
• /
• 2005

Objectives: To propose a risk-adjustment model from insurance claims data, and analyze the changes in cesarean section rates of healthcare organizations after adjusting for risk distribution. Methods: The study sample included delivery claims data from January to September, 2003. A risk-adjustment model was built using the 1st quarter data, and the 2nd and 3rd quarter data were used for a validation test. Patients' risk factors were adjusted using a logistic regression analysis. The c-statistic and Hosmer-Lemeshow test were used to evaluate the performance of the risk-adjustment model. Crude, predicted and risk-adjusted rates were calculated, and compared to analyze the effects of the adjustment. Results: Nine risk factors (malpresentation, eclampsia, malignancy, multiple pregnancies, problems in the placenta, previous Cesarean section, older mothers, bleeding and diabetes) were included in the final risk-adjustment model, and were found to have statistically significant effects on the mode of delivery. The c-statistic (0.78) and Hosmer-Lemeshow test ($x^2$=0.60, p=0.439) indicated a good model performance. After applying the 2nd and 3rd quarter data to the model, there were no differences in the c-statistic and Hosmer-Lemeshow $x^2$. Also, risk factor adjustment led to changes in the ranking of hospital Cesarean section rates, especially in tertiary and general hospitals. Conclusion: This study showed a model performance, using medical record abstracted data, was comparable to the results of previous studies. Insurance claims data can be used for identifying areas where risk factors should be adjusted. The changes in the ranking of hospital Cesarean section rates implied that crude rates can mislead people and therefore, the risk should be adjusted before the rates are released to the public. The proposed risk-adjustment model can be applied for the fair comparisons of the rates between hospitals.

• The Korean Journal of Community Living Science
• /
• v.15 no.3
• /
• pp.167-177
• /
• 2004

This study is designed to understand purchase motivation for garment depending on demographic factors among college students in Yanbian, China. Questionnaire was used for measurement tools to study the subject of the thesis. The main study was conducted against 450 college students from May 17 to June 5, 2001. The data for the study were analyzed using SAS PC program for frequency distribution, percentage, t -test, and one way ANOVA. The purchase motivation for garment are affected by demographic factors such as gender, average monthly household income, monthly expense for clothing. The result was showed as follows: A meaningful difference showed in 3 areas 'to try a new trend, impulsive buying at the store display, discount advertising' depending on the gender in terms of purchase motivations, and in all the three areas, male students showed a higher ranking. But in other motivation areas, no difference was noticed in terms of gender. In terms of purchase motivation based on monthly income, only one area 'impulse buying from a store display' showed a meaningful difference. Respondents with an average monthly household income above 2,000 yuan showed a higher tendency of 'impulse buying' compared to those with below 500 yuan or those with between 500-2,000yuan. Those with the average monthly household income below 500 yuan showed the lowest ranking in the impulse buying. In other areas of purchase motivation, average monthly household income was not an important element. A meaningful difference showed in 4 areas, 'to try a new trend, impulse buying from a store display, discount advertising', and 'for a change of mood' in the product motivation based on expense on clothing. Respondents with an average monthly expense for clothing above 100 yuan showed a higher ranking in all 4 areas than those with less than 100 yuan. In other areas, the average monthly clothing expense didn't give any impact.

• Journal of Labour Economics
• /
• v.40 no.2
• /
• pp.1-29
• /
• 2017

We examine the extent to which students' performance relative to peers affects their career choice. Specifically, we analyze the relationship between a student's mathematics ranking in his/her school and the likelihood of choosing Mathematics and Science track in high school. Using a panel dataset of students in Seoul, we measure a student's performance using two variables: absolute performance and relative performance. The former measures a student's performance relative to the entire sample, while the latter measures performance relative to the student's peers in the same school. After controlling for test scores and other characteristics, we find that the students with a poor relative ranking are 11 percentage points less likely to choose the Mathematics and Science track. Relative performance affects girls more greatly than boys. Although relative performance affects a student's self-efficacy and class participation, our accounting exercise suggests that this channel accounts for only 12 percent of the impact, implying that students may respond to the relative ranking mostly due to other factors, such as strategic consideration to perform well in college applications.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.1-8
• /
• 2019

Among the methods for extracting the most appropriate information from a large amount of log data, there is a method using inductive inference. In this paper, we use SVM (Support Vector Machine), which is an excellent classification method for inductive inference, in order to determine the ranking of intrusion logs in digital forensic analysis. For this purpose, the logs of the training log set are classified into intrusion logs and normal logs. The associated words are extracted from each classified set to generate a related word dictionary, and each log is expressed as a vector based on the generated dictionary. Next, the logs are learned using the SVM. We classify test logs into normal logs and intrusion logs by using the log set extracted through learning. Finally, the recommendation orders of intrusion logs are determined to recommend intrusion logs to the forensic analyst.

• Journal of the Korean Statistical Society
• /
• v.17 no.2
• /
• pp.61-80
• /
• 1988

For testing homogeneity of scale parameters aginst ordered alternatives, some nonparametric test statistics based on pairwise ranking method are proposed. The proposed tests are distribution-free. The asymptotic distributions of the proposed statistcs are also investigated. It is shown that the Pitman efficiencies of the proposed rank tests are the same as those of the corresponding two-sample rank tests in the scale problem. A small-sample Monte Carlo study is also performed. The results show that the proposed tests are robust and efficient.

• Journal of Korean Society for Quality Management
• /
• v.19 no.1
• /
• pp.16-27
• /
• 1991

Propose three rank tests based on different kinds of ranking methods for comparing several treatments with a control in a randomized block experiment. Monte Carlo power simulation study is examined in some small sample sizes and configurations to recommend a better test for applications.

• Journal of Korean Society for Quality Management
• /
• v.40 no.4
• /
• pp.497-512
• /
• 2012

Purpose: Most major countries have their own set of qualifications called national quality awards to measure the quality of companies and organizations. This study analyzes 3 different national quality awards and compare with the result from Korean quality awards conducted by Ministry of knowledge and Economy and Korea institute for advancement of technology. Methods: We tested 17 technoparks out of 18 technoparks in Korea and see how different the results can be depends on the value weights. We closely looked at each qualifications and tables of different countries' awards and compared with one used in Korea. Finally we proposed some suggestions to use not only domestic model but also international ones to be objective and add efficiency to organizations. Results: Depend on similarity of qualifications and weights, there were countries with different results and these caused score and ranking changes. Nevertheless, there was a comparison that did not make any changes on both score and ranking. Conclusion: We recognized the limitation that a standardized quality variation cannot be enough sources to test and analyze technoparks with different size and criteria. Integrating global standards and flow would be the first step to help grow technoparks and organizations placed in Korea in days to come.

• Journal of the Korean Graphic Arts Communication Society
• /
• v.29 no.2
• /
• pp.15-32
• /
• 2011

In printing, managing color means that how closely its color reproducts and printing supplier meets customers' requirements. When applying device profiles, it depends on properties of the devices. But color management of domestic digital prints is accomplished more scientifically and objectively than any other printing. According to this paper addresses a method of evaluating between proof prints and offset prints which are produced by identical date on the field. We evaluate two cases normal proof prints and domestic offset prints based on standardized color data analysis and subjective data analysis. We gathered objective data by measuring solid density, $CIEL^*a^*b^*$ and ${\Delta}E^{*_}{ab}$. Furthermore, we evaluated the offset prints and proof prints through human eyes to decide the ranking.

• KIISE Transactions on Computing Practices
• /
• v.21 no.2
• /
• pp.144-147
• /
• 2015

Question Answering (QA) services can provide exact answers to user questions written in natural language form. This research focuses on how to build a QA system for a specific domain area. Online and offline QA system architecture of targeted domain such as domain detection, question analysis, reasoning, information retrieval, filtering, answer extraction, re-ranking, and answer generation, as well as data preparation are presented herein. Test results with an official Frequently Asked Question (FAQ) set showed 68% accuracy of the top 1 and 77% accuracy of the top 5. The contribution of each part such as question analysis system, document search engine, knowledge graph engine and re-ranking module for achieving the final answer are also presented.

• ETRI Journal
• /
• v.45 no.3
• /
• pp.448-461
• /
• 2023

This article performs a detailed data scrutiny on a chronic kidney disease (CKD) dataset to select efficient instances and relevant features. Data relevancy is investigated using feature extraction, hybrid outlier detection, and handling of missing values. Data instances that do not influence the target are removed using data envelopment analysis to enable reduction of rows. Column reduction is achieved by ranking the attributes through feature selection methodologies, namely, extra-trees classifier, recursive feature elimination, chi-squared test, analysis of variance, and mutual information. These methodologies are ranked via Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) using weight optimization to identify the optimal features for model building from the CKD dataset to facilitate better prediction while diagnosing the severity of the disease. An efficient hybrid ensemble and novel similarity-based classifiers are built using the pruned dataset, and the results are thereafter compared with random forest, AdaBoost, naive Bayes, k-nearest neighbors, and support vector machines. The hybrid ensemble classifier yields a better prediction accuracy of 98.31% for the features selected by extra tree classifier (ETC), which is ranked as the best by TOPSIS.