• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.323-325
• /
• 2021

In this article, we have improved the prediction of hypertension detection using the feature selection method for the Korean national health data named by the KNHANES database. The study identified a variety of risk factors associated with chronic hypertension. The paper is divided into two modules. The first of these is a data pre-processing step that uses a factor analysis (FA) based feature selection method from the dataset. The next module applies a predictive analysis step to detect and predict hypertension risk prediction. In this study, we compare the mean standard error (MSE), F1-score, and area under the ROC curve (AUC) for each classification model. The test results show that the proposed FIFA-OE-NB algorithm has an MSE, F1-score, and AUC outcomes 0.259, 0.460, and 64.70%, respectively. These results demonstrate that the proposed FIFA-OE method outperforms other models for hypertension risk predictions.

• 한국지리정보학회지
• /
• 제16권2호
• /
• pp.101-113
• /
• 2013

좋은 도시경관을 만들기 위해서는 경관분석의 역할이 매우 중요하다. 경관분석은 여러 단계를 거치게 되는데, 그 중 가장 기본이며 첫 단계가 조망점을 선정하는 것이다. 그동안 조망점 선정은 정성적인 기준에 따라 전문가들의 주관적 판단에 의해 선정되었는데, 이 과정에서 질 높은 조망점이 누락되거나 무시되기도 하였다. 본 연구의 목적은 수치지형도 등의 데이터와 GIS 기술을 활용하여 경관분석의 기본이 되는 조망점 선정 방법을 제시하는 것이다. 본 연구에서는 주요 도로의 교차점, 쇼핑센터 등 주요 시설, 교량 및 육교, 오픈스페이스 등 조망점으로 선정될 수 있는 기준에 해당하는 지점을 각종 자료 및 GIS 기술을 이용하여 추출하는 방법을 제시하였다. 이를 위하여 도로대장의 도형정보와 속성정보 등 데이터와 질의(query), 중첩(overlay) 등의 기능을 적용하였다. 이와 같은 과정을 거쳐 선정된 예비조망점에 대하여 수치지형도의 표고점을 이용한 수치표고모델(DEM)을 구축한 후 시뮬레이션 과정을 통하여 조망점에서의 조망대상에 대한 차폐유무를 분석하였다. 분석 결과, 기존의 방법에 비해 적은 시간과 비용을 투입하여 체계적인 조망점을 선정할 수 있었고, 선정된 각 조망점의 차폐도도 자동으로 분석할 수 있었다. 따라서, 1차적으로 GIS 기법을 이용한 조망점 분석 이후 전문가의 현장조사를 수반한다면 조망점 선정 및 분석 결과가 향상될 것으로 기대된다.

• Journal of the Korean Data and Information Science Society
• /
• 제28권5호
• /
• pp.1077-1085
• /
• 2017

본 연구에서는 한국노동연구원의 "2015년 한국노동패널조사 (KLIPS)" 자료를 활용하여 국내 여성의 임금 결정요인을 분석하기 한다. 일반적으로 임금 자료는 랜덤 추출이 불가능하기 때문에 분석하기가 쉽지 않다. 표본 선택 편의 (sampling bias)가 있는 자료를 분석하는 방법으로 Heckman 표본 선택 모형이 가장 널리 알려져 있다. Heckman은 크게 두 가지 모형을 제안했는데, 그 중 하나는 최대 우도 방법을 이용하는 것이고, 다른 하나는 2단계 표본 선택 모형이다. 이 중 Heckman 2단계 표본 선택 모형은 주된 결과 모형 (outcome model)과 경제 활동 여부를 결정짓는 선택 모형 (selection model)을 포함한 모형으로써, 이 모형이 최대 우도 방법을 이용한 모형에 비해 이변수 오차의 정규분포 가정에 덜 민감하다고 알려져 있다. 그럼에도 불구하고 이변수 오차에 대한 정규 분포 가정은 꽤 강한 가정이라고 볼 수 있는데, 최근에 이 모형의 단점을 보완하는 모형으로 Marchenko와 Genton (2012)의Heckman 표본 선택 t 모형이 제시되었다. Heckman 2단계 모형과 Heckman 표본 선택 t 모형을 이용하여 국내 여성의 임금 결정 요인을 분석하고 비교하도록 한다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제21권3호
• /
• pp.320-324
• /
• 2008

We investigated the effect of ESR gene-based selection on an improvement of litter size in the herds in real (non-experimental) conditions. The pigs were selected for three years. In the tested population the pigs were mated according to a breeding scheme where the individuals with at least one ESR-B allele were preferred in the selection. In the control group (CP; n = 140) the pigs were mated just according to a breeding scheme without knowledge of the ESR genotype. We observed a significant increase in litter size (total number of born, number of born alive and number of weaned piglets per litter) in the final tested ESR-selected population (LP; n = 184) and an insignificant increase in CP as compared with the original population (OP; n = 155). After the selection we could observe a significant increase in the frequency of allele B in LP. Frequency of the genotypes AB and BB increased in both LP and CP; the distribution of the genotypes changed significantly only in LP. An association analysis of the ESR gene effects on reproductive traits in LP showed no significant differences between the genotypes. The results of our study suggest that ESR gene-based selection can be successful also in small herds, under real (non-experimental) conditions with a respect for general breeding principles and limitations and during a short period. An examination of a larger sample population as well as an analysis of selection consequences on other traits (meat and carcass quality) could bring a more conclusive evaluation of ESR-based selection. Nevertheless, the results are encouraging especially for small breeding farms taking a perspective of better litter size improvement.

• 한국의류산업학회지
• /
• 제8권1호
• /
• pp.55-63
• /
• 2006

The purpose of this study examines difference of cunsumer satisfaction by expected selection attribute of wedding dress and bride make-up. It is looking into influencing variables in consumer satisfaction, and is supplying real information to wedding market. The subjects of this study were 241 women getting married within one year of Daegu. The data were analyzed by using frequency, mean, the standard deviation, ANOVA, duncan test, correlation analysis, cross-analysis. Conclusion obtained from the results and discussions of the research are as follows: 1) In the attribution of expected selection according to occupation, blue-collar women are concerned about cost factor but they are low content with the perceptual satisfaction factor. 2) When selecting places for wedding dresses and wedding make-up, in the attribute of expected selection about the number of visiting the shop for wedding dresses and make-up, people who visit over 7 dress shops are highly interested in the appearance fancy degree, advertisements effect factors. But they are concerned with advertisements effect factor, cost factor in the perceptual satisfaction factor about wedding dress. Besides, the more they get chances to visit many dress shops, they are concerned with advertisements, cost factors, and satisfied with both the effect of advertisements and appearance factor degree in the attribution of expected selection about wedding make-up. 3) In the attribution of expected selection about the cost of wedding dresses and make-up, the lower the cost paid, people are interested in price factor. Furthermore, in the attribution of expected selection about the wedding dress and make-up cost, those who spent over 150 million Won are interested in quality, others influence factor, and the case of less than 120~150 million Won, people are highly interested in the appearance fancy degree, advertisements effect factor, and the women are satisfied with the advertisement effect factor in the perceptual satisfaction factor of wedding make-up.

• Journal of the Korean Data and Information Science Society
• /
• 제25권1호
• /
• pp.187-194
• /
• 2014

We consider a Bayesian nonignorable selection model to accommodate the selection bias. Markov chain Monte Carlo methods is known to be very useful to fit the nonignorable selection model. However, sensitivity to prior assumptions on parameters for selection mechanism is a potential problem. To quantify the sensitivity to prior assumption, the deviance information criterion and the conditional predictive ordinate are used to compare the goodness-of-fit under two different prior specifications. It turns out that the 'MLE' prior gives better fit than the 'uniform' prior in viewpoints of goodness-of-fit measures.

• Journal of Construction Engineering and Project Management
• /
• 제7권2호
• /
• pp.1-10
• /
• 2017

The Clonal Selection Algorithm (CSA) is an algorithm inspired by the human immune system mechanism. In CSA, several parameters needs to be optimized by large amount of sensitivity analysis for the optimal results. They limit the accuracy of the results due to the uncertainty and subjectivity. Adaptive Clonal Selection (ACS), a modified version of CSA, is developed as an algorithm without controls by pre-defined parameters in terms of selection process and mutation strength. In this paper, we discuss the ACS in detail and present its implementation in construction site utilization planning (CSUP). When applied to a developed model published in research literature, it proves that the ACS are capable of searching the optimal layout of temporary facilities on construction site based on the result of objective function, especially when the parameterization process is considered. Although the ACS still needs some improvements, obtaining a promising result when working on a same case study computed by Genetic Algorithm and Electimze algorithm prove its potential in solving more complex construction optimization problems in the future.

• 한국의류학회지
• /
• 제12권2호
• /
• pp.159-167
• /
• 1988

The purpose of this study was to investigate the relationship between Self-concept, Personal variables and Clothing selection behavior. Self-concept was measured with Choi Jung Hun's 'Perceptual Orientation Scale' and Clothing selection behavior scale was prepared for this study. The questionnaire were completed by 389 women in Taegu. Statistical analysis was performed using F-test, Scheffe's test. The results were as follows; 1. There was significant relationship between Self-concept and Clothing selection behavior. (individuality, conformity, economy, modesty). 2. There was significant difference in clothing selection behavior variables according to age. 3. There was significant difference in individuality and economy according to marital status. 4. There was significant difference in individuality, economy and modesty according to education level. 5. There was significant difference in clothing selection behavior variables according to monthly clothing expenses.

• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.195-205
• /
• 2005

In decision tree analysis, C4.5 and CART algorithm have some problems of computational complexity and bias on variable selection. But QUEST algorithm solves these problems by dividing the step of variable selection and split point selection. When input variables are continuous, QUEST algorithm uses ANOVA F-test under the assumption of normality and homogeneity of variances. In this paper, we investigate the influence of violation of normality assumption and effect of the transformation of variables in the QUEST algorithm. In the simulation study, we obtained the empirical powers of variable selection and the empirical bias of variable selection after transformation of variables having various type of underlying distributions.

• 한국건설관리학회논문집
• /
• 제16권1호
• /
• pp.82-91
• /
• 2015

최근 전 세계적으로 친환경 및 저탄소에 대한 관심이 증가하고 있으며, 이에 따라 대표적인 친환경연료인 천연가스의 사용량이 급속도로 증가하고 있다. 특히 국내에서도 산업용, 발전용, 가정용으로 사용되는 천연가스 수요증가에 대비하여 도입 물량을 확대하기 위한 전략을 수행하고 있으며, 이에 따른 LNG 저장용량의 확보를 위해 저장기지 증설을 계획하고 있다. 그러나 기존의 LNG 저장기지의 입지선정은 기업의 내부적인 절차나 용역을 통해 수행되어 LNG 및 LNG 저장기지의 특성을 반영하는 데는 미흡하였다. 또한 해외에서도 LNG 저장기지의 입지선정과 관련하여 주요 요인들에 대한 연구가 수행되고 있으나 프로세스 및 모델 등 체계적인 분석은 미흡한 상황이다. 따라서 본 연구에서는 LNG 저장기지의 특성을 고려한 입지선정 모델을 구축하고자 한다. 이를 위해 관련기업의 과거 사례를 분석하여 저장기지 입지선정 과정에서 요구되는 요인들과 기존에 연구되어온 플랜트시설, 공장, 산업단지, 관청청사 등의 입지선정에 대한 요인들을 취합하여 전문가 인터뷰를 실시하였고 최종적으로 47개의 입지선정요인을 도출하였다. 이후 기 수행된 5개 지역(PT지역, IC지역, TY지역, SC지역, BR지역)의 사례에 대한 설문을 기반으로 요인분석, 다중회귀분석을 통하여 지역별 입지선정에 대한 우선순위를 도출하였고 이를 토대로 LNG 저장기지 후보지에 대한 입지선정 모델의 활용 가능성을 검토하였다. 향후 LNG 저장기지의 추가 건설과정에서 본 연구를 기초자료로서 활용한다면, 보다 효과적이고 체계적인 입지선정이 가능할 것으로 기대된다.