• The Korean Journal of Applied Statistics
• /
• v.22 no.3
• /
• pp.531-539
• /
• 2009

Logistic regression is widely used as a datamining technique for the customer relationship management. The maximum likelihood estimator has highly inflated variance when multicollinearity exists among the regressors, and it is not robust against outliers. Thus we propose the robust principal components logistic regression to deal with both multicollinearity and outlier problem. A procedure is suggested for the selection of principal components, which is based on the condition index. When a condition index is larger than the cutoff value obtained from the model constructed on the basis of the conjoint analysis, the corresponding principal component is removed from the logistic model. In addition, we employ an algorithm for the robust estimation, which strives to dampen the effect of outliers by applying the appropriate weights and factors to the leverage points and vertical outliers identified by the V-mask type criterion. The Monte Carlo simulation results indicate that the proposed procedure yields higher rate of correct classification than the existing method.

• Journal of Korea Spatial Information System Society
• /
• v.7 no.3 s.15
• /
• pp.13-24
• /
• 2005

This paper proposes a model for the location information grading policy. Location information grading policy is that government would give a public notice of criterion for location information quality, and location information operators should elucidate the grade of provided location information. As the quality grade of services is provided in the 'terms of use' with the detailed quality specification and quality range, the user can understand the quality of location information, and can select more suitable service for their purposes. Additionally, the qualiy grading policy can enforce the location information operators and location service providers to support more stable and reliable service quality. In this paper, we propose the model of location information grades, the typical classification of quality, and the realitic policy applying model. Recent enacted 'Location Information Law' are expected to activate korean information market. In this situation, proposed location information grading policy take a role in protecting location privacy and improving the reliability of location information services.

• Korean Journal of Community Nutrition
• /
• v.18 no.5
• /
• pp.525-538
• /
• 2013

The purpose of this study was to provide information on obesity assessment for Koreans. Among total of 1012 research papers enlisted in the Korean J Community Nutrition form 1996 to 2011, 248 articles were examined in which subjects were divided into more than 2 groups by obesity rate. About the method of anthropometric data collection, more than half of the research papers examined 52.5% and 28.7% of studies utilized the directly measured data and self-described data, respectively. About the utilization of obesity assessment methods, indirect methods of weight-height index (BMI, BMI percentile, and R$\ddot{o}$hrer index) and PIBW (WLR, Broca index, and KDA) were 62.4% and 23.2%, respectably, and the direct method of percent body fat assessment was only 9.3%. The most frequently utilized methods were WLR in under primary and primary school children, and BMI in the middle and high school students and in adults. For primary school students, WLR was the most frequently utilized method up to 2007, but it changed to BMI percentile afterward. Broca Index was no longer utilized since 2008. There were no articles utilizing BMI percentile and R$\ddot{o}$hrer index for obesity assessment in adults. Criteria for obesity assessment were not consistent among research papers: for example, % body fat, 19~40%; BMI, 20~30; BMI percentile, 85th or 95th. In the case of PIBW, 120% of ideal weight was the most frequently utilized criterion for obesity. Based on these findings, we suggest that proper methods and criteria of obesity assessment for each age group should be determined and proclaimed.

• Economic and Environmental Geology
• /
• v.46 no.6
• /
• pp.545-559
• /
• 2013

Until now, research achievements of groundwater such as groundwater to depth distribution, usage, the available amount of development, water quality have been written in the watershed units($25{\sim}250km^2$). However, complex topography and geology, and the rivers of our country does not fit. And a clear management standards have not been able to present measures in groundwater quantity, water quality management such as rainfall, groundwater, utilization, water quality, pollution, etc. Therefore, in this study, the classification criterion of subwatershed unit($2.5{\sim}25km^2$), which is suitable for topography and geology of Korea, for rainfall-rating, groundwater level-rating, groundwater pollution-rating, groundwater quality-rating presented and proved its efficiency by applying in Hampyeong-Gun area.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.17 no.1
• /
• pp.190-196
• /
• 2009

This study proposes a structural design method for the upper control arm installed at the rear side of a SUV. The weight of control arm can be reduced by applying the design and material technologies. In this research, the former includes optimization technology, and the latter the technologies for selecting aluminum as a steel-substitute material. Strength assessment is the most important design criterion in the structural design of a control arm. At the proto design stage of a new control arm, FE (finite element) analysis is often utilized to predict its strength. This study considers the static strength in the optimization process. The inertia relief method for FE analysis is utilized to simulate the static loading conditions. According to the classification of structural optimization, the structural design of a control arm is included in the category of shape optimization. In this study, the kriging interpolation method is adopted to obtain the minimum weight satisfying the strength constraint. Optimum designs are obtained by ANSYS WORKBENCH and the in-house program, EXCEL-kriging program. The optimum results determined from the in-house program are compared with those of ANSYS WORKBENCH.

• Journal of Fashion Business
• /
• v.20 no.2
• /
• pp.181-196
• /
• 2016

This study intended to suggest criteria for selection of subjects by lower body shape types necessary for evaluating slacks. For this, the characteristics were examined by lower body parts which would influence the fit of slacks on 3D human body shape data of the front and sides of the lower body for lower body shaping. The frequency of subjects by lower body shape types and the boundary points for discrimination of each type were suggested so that they could be available in selecting subjects. Using the data from Size Korea(2004), indirect measurement values measured on the front and sides of the lower body among 3D human body shape data of 175 subjects were analyzed. Their height, waist, and hip circumference fell under the range of standard deviation based on the mean of women aged 18~24 years, and then lower body shaping was conducted by combining the front and side shapes of the lower body. The front of the lower body was classified into four sections: average waist/average hip type(F1), average waist/narrow hip tyle(F2), narrow waist/narrow hip type(F3) and narrow waist/wide hip type(F4) and the sides of the lower body were divided into four sections: average abdomen/average hip type(S1), flat abdomen/average hip type(S2), average abdomen/protrude hip type(S3)and round abdomen/flat hip type(S4), and thus total 16 lower body types were created by cross analysis. Besides, discriminant analysis suggested the boundary points for each shape type of the front and sides of the lower body as a criterion for deciding lower body shape type of each subject

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.2
• /
• pp.137-148
• /
• 2011

In this paper, we systematically investigate regional differences of the dead due to injuries in cities, towns and counties about transportation accidents, suicides and fall accidents, which have recently been an important issue of health problems in Korea, The data are from the Annual Report on the Cause of Death Statistics in Korea in 2008. They include the deaths over the age 19 from transportation accidents, suicides and fall accidents with the criterion of the International Statistical Classification of Diseases. Poisson HGLM is applied to estimate the mortality rate under the assumption that the number of deaths follow a Poisson distribution, by considering regions as random effects and by adjusting age, sex and standardized residence tax as fixed effects. Using the results of random effects prediction, the regional differences in cities, counties and towns are marked in disease mapping. The results showed that there were significant regional differences of mortality rates for transportation accidents and suicides, but no significant differences for fall accidents.

• The Journal of the Korea Contents Association
• /
• v.12 no.11
• /
• pp.39-48
• /
• 2012

It is well known that software testing amounts for a significant portion of software development cost. In order to reduce the cost of software testing. a lot of researches on automated test data generation have been performed. Sophisticated tools for performing symbolic execution or solving a system of path constraints are required to support automated test data generation. Developing or purchasing those tools leads to another factor of increasing the cost involving software testing. In this paper, we propose a dynamic test data generation approach that does not depend on symbolic execution or constraint solving at all. The proposed approach extends Korel's path-oriented method to satisfy the branch coverage criterion effectively. We conducted an experiment to evaluate the effectiveness of the proposed technique with a triangle classification program to show that branch coverage can be easily achieved.

• Journal of Acupuncture Research
• /
• v.30 no.3
• /
• pp.109-115
• /
• 2013

Objectives : This study was aimed to find the motivation and satisfaction of subjects who participated in acupuncture and moxibustion clinical trials and to suggest ways to improve clinical trial management. Methods : We conducted a survey and collected the data of 233 subjects of acupuncture and moxibustion clinical trials from 6 university hospitals. Results : For all 233 subjects, the mean of the total satisfaction score was 4.07 (perfect score 5.0), and the score was particularly high in aspects of the investigator and clinical research coordinator, but low in the site and sponsor aspects. There were no differences in the satisfaction degree with general characteristics. The satisfaction degree of the acupuncture and moxibustion clinical trials was strong positive correlated with participation benefits(r=0.595) and easy contact with the research team (r=0.500). Conclusions : Most subjects showed positive perception and interest in the clinical trials. Future studies should investigate how personal factors influence the satisfaction of subjects, should identify a specific classification criterion, and create a standardized protocol for the clinical trials.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.23 no.1
• /
• pp.111-122
• /
• 1995

12 populatiDns of Ch indicum L. and it's variation antogenous in Tforea were selected throughout the country and their morphological characteristics were studied in this paper. The results are as follows. 1) The mean of the plant height is 97.3cm. Ch. indicunl is 76.5cm and Ch. boreale is 116. 6cm in height. The number of nodes is 58 on the average. The color of the stem is generally black-red in the case of Ch. indicum and is green for Ch. boreale. 2) The leaves of Ch. inficum are smaller In size and thicker than chose of Ch. boreale, and it has a stipule. This fact enables us to distinguish one species from the other, however, the Gyukpo-population of Ch. indicum has a large leaf, and Byunsan-population of Ch. boreale has a stipule so that the above fact is not an exact criterion on which to base one classification. 3) The size of the capitulum is 24.9mm for Ch. indicum and 15.6mm in the case of Ch. boreaje so that the difference between species is admitted. It was confirmed hat the size of the capitulum has positive correlation to that of the pedicel and has egative correlation to the number of flowers, so that of inflorescence, Ch. indicumhas corymb, and Ch. boreale has crowed corymb or umbel-type.The number of ligulates is 19 on the average, which no the difference being founnd between species and populations.4) The cluster analysis of morphological characteristics showed that Ch. indicum wasdivided into two groups, which was shown thought to be due to the difference of speciesbetween Ch. indicum and Ch. boreale. However, it could not be shown that their morphological characteristics explain a geographical relationship between populations