• 한국응용과학기술학회지
• /
• 제25권4호
• /
• pp.438-445
• /
• 2008

For recipe prediction of colorant proportion showing nonlinear behavior, we modeled the effects of colorant proportion of basic colors on the target colors and predicted colorant proportion necessary for making target colors. First, colorant proportion of basic colors and color information indicated by the instrument was applied by a linear model and a multi-layer perceptrons model with back-propagation learning method. However, satisfactory results were not obtained because of nonlinear property of colors. Thus, in this study the neuro-fuzzy model with merit of artificial neural networks and fuzzy systems was presented. The proposed model was trained with test data and colorant proportion was predicted. The effectiveness of the proposed model was verified by evaluation of color difference(${\Delta}E$).

• Animal cells and systems
• /
• 제6권4호
• /
• pp.293-299
• /
• 2002

We investigated seasonal variation of diurnal activity patterns in Jeiu ponies (Equus caballus). The data were collected from a semi-natural herd during daylight hours from March to September 1998. Over all months the mean proportion of time spent grazing was 83.7% ( $\pm$29.7 S.D.): that resting was 15.7$\pm$17%, grooming 0.7$\pm$1.2%, and moving 0.56$\pm$5.4%, Activity patterns varied with the seasons. The mean proportion of grazing mares peaked during late winter and early spring (March), when there was not enough food, and declined during other seasons (between May and September), when food was abundant. The mean proportion of mares that were resting and mutual grooming peaked in spring (April and May), at which time the weather was warm and food became abundant. During other seasons when grasses started to grow and the weather was mildly cool, these activities were less common. The mean proportion of mares that were moving peaked in June, although it was a small proportion of the total activity. For those mares without foals, resting periods were longer in older mares than in younger mares, but for the mares with foals this pattern was not evident. From these data, we hypothesize that the dominance hierarchy of the mares effects the diurnal activity pattern.

• 한국의류산업학회지
• /
• 제4권6호
• /
• pp.534-543
• /
• 2002

The purpose of this study was to identify the body proportions of girls in middle and high school who have been a major sales target of clothing manufacturers and thus to provide the basic information on the proportion of human body, which is essential for manufacturer to produce clothing and for individual consumers to select right apparels to wear and purchase. The subjects were 133 girls in middle school and 141 girls in high school. As measuring instruments, the Martin's anthropometer, tapelines, and belts for marking waistline were used. In order to analyze the data, t-test and factor analysis were used. The measurement of vertical distance between vertex and measuring points, showed a significant difference between the high school group and middle school group in every item. In the comparison of eight head proportion between the two groups, middle school students had 7.2 head proportion while girls in high school had 6.9 head proportion in detail. The factor analysis of major divided body parts of the two groups indicated that waist-hip and hip-crotch, constituting lower trunk together, vertex-apex nasi, and apex nasi-back neck point, constituting head together, back neck point-spina scapula and spina scapula-waist, constituting posterior of upper body together, and mentum-bust point and bust point-waist, constituting anterior of upper body together, were in inverse proportion.

• Journal of Preventive Medicine and Public Health
• /
• 제37권4호
• /
• pp.345-352
• /
• 2004

Objectives : Previous studies showed that death certification by physicians was an important predictor to improve the quality of death certificate data in South Korea. This study was conducted to examine the proportion of death certificates issued by physicians and associated factors in South Korea from 1990 to 2002. Methods : Data from 3,110,883 death certificates issued between 1990 and 2002, available to the public from the National Statistical Office of Korea, were used to calculate the proportion of death certificates issued by physicians and to examine associated factors with logistic regression analysis. Results : The overall proportion of death certificates issued by physicians increased from 44.6% in 1990 to 77.6% in 2002 (mean: 63.5%). However, the proportion was greatly influenced by the deceased's age. In 2002, more than 90% of the deceased aged 51 or less were certified by physicians. A higher proportion was found among deceased who had tertiary education (college or higher) living in more developed urban areas. Conclusion : The information regarding the cause of death for younger, well-educated deceased in urban areas of South Korea may show a higher level of accuracy. Epidemiologic research using information on causes of death may well benefit from the continually increasing proportion of death certificates issued by physicians in the future in South Korea.

• The Journal of Advanced Prosthodontics
• /
• 제4권2호
• /
• pp.72-75
• /
• 2012

PURPOSE. This study evaluated the existence of golden proportion between the widths of the maxillary and mandibular anterior teeth in Indian population. MATERIALS AND METHODS. The clinical tooth width measurements were recorded with the digital vernier calipers on 576 patients of both sexes in the age group of 21 - 30 years. Flexible ruler was used to determine the width of maxillary and mandibular anterior teeth on the patients by the same operator. The data obtained was statistically analyzed using paired student t-test (${\alpha}$=.05). RESULTS. The golden proportion was not found between the width of the right central and lateral incisors in 53% of women and 47% of men. The results revealed the golden percentage was rather inconstant in terms of relative tooth width. CONCLUSION. The golden proportion is an inappropriate method to relate the successive widths of the maxillary anterior teeth in Indian population.

• 가정과삶의질연구
• /
• 제17권4호
• /
• pp.193-206
• /
• 1999

This study analyzed the household portfolio according to wealth levels using a survey data of 1997 Korea Household Panel Study. The major findings of this study are as follows: (1) A household in high wealth level has invested relatively large proportion of his asset into real estate (2) A household in middle wealth level has invested relatively large proportion of his asset into risky financial asset(3) A household in low wealth level has invested relatively large proportion of his asset into secure financial asset. These findings accorded with risky pyramid model.

• Journal of the Korean Data and Information Science Society
• /
• 제26권4호
• /
• pp.857-864
• /
• 2015

데이터 마이닝은 다양한 형태의 방대한 데이터 집합으로부터 보이지 않는 지식이나 새로운 법칙을 발견한 후, 이를 바탕으로 의사결정 등을 위한 정보로 활용하고자 하는 것이다. 데이터 마이닝 기법중의 하나인 군집 분석은 거리 또는 유사성 측도를 이용하여 집단을 분류하고, 구분된 각 집단의 특성을 파악하기 위한 기법이다. 본 논문에서는 주변 확률이 일부 포함된 확률적 흥미도 측도 기반의 유사성 측도들인 Peirce I, Peirce II, Cole I, Cole II, 그리고 이들을 응용한 Park I 및 Park II에 대한 대소 관계를 수식의 증명뿐만 아니라 예제 데이터에 의해서도 규명하였다. 그 결과, Cole I과 Cole II의 측도를 동시에 고려한 Loevinger 측도가 기존의 측도들 중에서는 상한이 되나 Park I 및 Park II를 함께 고려했을 경우에는 동시발생비율, 동시 비발생비율, 그리고 두 가지 형태의 불일치비율의 크기에 따라 변한다는 사실을 확인하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제28권3호
• /
• pp.669-684
• /
• 2017

Mediation models are widely used in many fields of research and have recently gained attention in epidemiology. The mediation proportion is a standard measure to evaluate what part of the total exposure effect on an outcome may be explained by a particular mediator and to examine how important that pathway is relative to the overall exposure effect. A common question is how large a sample size is needed to achieve high statistical power or, equivalently, what magnitude of effect can be detected. Current power and sample size calculations for mediation analysis are limited and additional research is needed. We therefore propose a computer-intensive power analysis using the mediation proportion. We conduct simulation studies to calculate statistical powers and sample sizes. And then, we illustrate our power analysis using an example from the Adult Health Study of atomic-bomb survivors and demonstrate that the method is relatively straightforward to understand and compute.

• 보건행정학회지
• /
• 제32권1호
• /
• pp.107-112
• /
• 2022

Catastrophic healthcare expenditure refers to out-of-pocket spending for healthcare exceeding a certain proportion of a household's income and can lead to subsequent impoverishment. The aim of this study was to investigate the proportion of South Korean households that experienced catastrophic healthcare expenditure between 2006 and 2020 using available data from the National Survey of Tax and Benefit (NaSTaB), Korea Health Panel (KHP), and Households Income and Expenditure Survey (HIES). Trend test was used to analyze the proportion of household with catastrophic healthcare expenditure. In the NaSTaB 2020 data, households who experienced catastrophic health expenditure was 1.73%. Trend analysis was significant with the decreasing trend (annual percentage change [APC], -5.55; p<0.0001) in the proportion of households with the catastrophic health expenditure. Also, in the 2018 KHP and the 2016 HIES, households who experienced catastrophic health expenditure was 2.21% and 2.92% respectively. In contrast, the trend was significantly increased in the KHP (APC, 0.55; p<0.0001) and the HIES (APC, 1.43; p<0.0001). Therefore, the findings suggest the need to strengthen public health care financial support and monitor catastrophic healthcare expenditures, especially for low-income group.

• 응용통계연구
• /
• 제15권1호
• /
• pp.33-43
• /
• 2002

범주형 데이터의 결측치 추정을 위하여 최빈 범주법, 로지스틱 회귀분석, 연관규칙과 같은 다양한 방법이 연구되어 왔다. 본 연구에서는 이러한 방법의 추정 값을 결합하는 신경망 융합과 투표융합 방법을 제안하고 이의 성능을 시뮬레이션을 이용하여 비교하였다. 실험에 사용된 데이터의 특성을 나타내는 인자로는 (1) 입출력 변수간의 연결함수, (2) 데이터의 크기, (3) 노이즈의 크기 (4) 결측치의 비율, (5) 결측발생 함수를 사용하였다. 분석결과는 다음과 같다. 데이터의 크기가 작고 결측 발생 비율이 높으면 최빈 범주법, 연관규칙, 신경망 융합의 성능이 높게 나타났으며 데이터의 크기가 작고 결측발생 확률이 결측이 안된 나머지 변수에 높은 의존관계가 있으면 로지스틱 회귀분석, 신경망 융합의 성능이 높게 나타났다. 데이터의 크기가 크고, 결측치의 비율이 낮으면서, 노이즈가 크고 결측발생 확률이 결측이 안된 나머지 변수에 높은 의존관계가 있으면 신경망 융합의 성능이 높게 나타났다.