QFD에서 군집화방법에 의한 HOQ구조의 재조정

이성석(Rhee, Sung-Suk) 서원대학교 미래창조연구원 2007 과학과 문화 Vol.4 No.4

상태공간모형에 의한 시계열분석에 관한 연구

이성석(Rhee, Sung Suk) 서원대학교 미래창조연구원 1993 응용과학연구 Vol.2 No.1

문화재 직지의 인지도 및 이미지의 측정방법과 제고방안

이성석(Rhee, Sung-Suk) 서원대학교 직지문화산업연구소 2015 湖西文化論叢 Vol.24 No.-

The purpose of this paper is to help any project about Jikji. The various methods measuring the degree of the awareness and the image of the object are reviewed, and the proposed alternatives by many researchers to improve the degree of the awareness and the image of cultural property, Jikji, was discussed. Jikji has been registered as a worldwide cultural property at UNESCO in 2001, and it has been known as a remarkable symbol of Chongju or Chungbuk area. There are a lot of methods to measure the degree of the awareness and the image of an object. They are classified as the overall method and the attribute-based method. The results of this study are as follows. 1. The degree of the awareness and the image of Jikji is high in Chongju or Chungbuk area, but low the outside of that area. 2. There are a lot of various methods to measure the degree of the awareness and the image of Jikji. They are classified as the overall method and the attribute-based method. As the attribute-based method, multi-attribute attitude model, semantic differential method, multi-dimensional scaling method, and importance-performance analysis were reviewed. 3. A lot of alternatives to level up the degree of the awareness and the image of Jikji by many researchers were discussed. 4. The periodical and systematical survey to investigate the awareness and the image of Jikji is proposed.

실시간 에볼라 바이러스 전염병 모형의 전염확률분포추정

최일수,이성석,Choi, Ilsu,Rhee, Sung-Suk 한국데이터정보과학회 2015 한국데이터정보과학회지 Vol.26 No.1

The epidemic is seemed to be extremely difficult for accurate predictions. The new models have been suggested that show quite different results. The basic reproductive number of epidemic for consequent time intervals are estimated based on stochastic processes. In this paper, we proposed a transmission distribution estimation for Ebola virus disease epidemic model. This estimation can be easier to obtain in real time which is useful for informing an appropriate public health response to the outbreak. Finally, we implement our proposed method with data from Guinea Ebola disease outbreak. 전염병은 기초 감염 재생산 수가 시간에 따라 달라져서 상황을 관리하기 어렵기 때문에 확산을 통제하기 어려울 뿐 아니라 정확한 예측은 더욱 어렵다고 알려져 있다. 최근에 많은 모형들이 새롭게 제시되고 있으며 그에 따라 현저하게 다른 결과가 도출되고 있다. 연속된 시간에서 기초 감염 재생산 수는 일반적으로 확률과정이론이 적용되고 있다. 본 논문에서는 에볼라 바이러스 전염병 모형에서 전염확률분포의 추정 방법을 제시하였다. 이 방법은 대규모 전염병 발생에서 실시간 추정을 가능하게 함으로 적절한 질병관리를 용이하게 한다. 기니에서 발생한 에볼라 바이러스 자료를 제시한 방법으로 분석하였다.

의사결정나무모형을 사용한 성인 생애주기별 취업 영향요인 분석

곽민정,이성석,Kwak, Minjung,Rhee, Sung-Suk 한국데이터정보과학회 2016 한국데이터정보과학회지 Vol.27 No.6

세계적으로 유가하락과 더불어 경제가 침체되면서 우리나라도 저성장의 기조를 보이고 있고, 노동 시장에서는 취업난이 가중되고 있으므로 취업영향 요인을 파악하여 적절한 취업 정책을 수립하는 것이 절실한 현실이다. 따라서 본 연구에서는 제17차년도 노동패널자료를 사용하여 취업에 영향을 미치는 요인을 파악하고자 한다. 성인 생애주기는 청년기, 중장년기와 노년기로 구분하였으며 취업에 영향을 미치는 요인으로 인구통계학적 변수, 직업관련 변수 그리고 건강관련 변수를 고려하였다. 의사결정나무분석을 사용하여 분석한 결과 청년기에는 학력이 가장 중요한 요인이었으며, 중장년기에는 가장 중요한 요인이 성별이었고, 남성의 경우 건강상태, 여성의 경우, 직업훈련경험, 연령, 건강상태의 순으로 나타났다. 노년기에도 성별이 가장 중요한 요인이었고 그 다음으로 건강상태, 학력 등의 순으로 나타났다. Due to global economy recession with low oil price, Korea has stepped into a stage of sluggish development, and the unemployment has become a major issue. Hence, we study various risk factors influencing on unemployment using the Korean labor and income panel data of 2014. We first divide the adult life cycle into three categories, such as young adult, adult, and old adult. Then we consider demographic variables, occupational variables and health related variables as risk factors. The decision tree models have shown that education and gender are the most important factors respectively in young adult group and in adult group. Gender, health status, and education are influential factors in old adult group.

내향성 발톱의 교정을 평가하기 위한 발톱 C커브 각도 계측 방법의 제안

방효진 ( Hyo Jin Bang ),이성석 ( Sung Suk Rhee ) 한국미용학회 2011 한국미용학회지 Vol.17 No.4

In this paper, we proposed a new method to measure the C-curve angle of a toenail to evaluate an ingrowing toenail, and analyzed the measured angle of C-curve of toenail to find the related variables, such as, self-judged angle, presence of an ingrowing toenail, walking style, type of shoes, height of heel, the experience of foot care, education and occupation. Basically, in the study of ingrowing toenail, it is important for the C-curve angle of toenail to be a tool of measuring the degree of refraction of toenail, and of evaluating an ingrowing toenail. The survey was performed with questionnaires and the picture of the toenail of respondents were taken. The data from 197 respondents who are female students or female customers of the nailcare shop were analyzed by frequencies, t-test, and F-test (ANOVA). The results of this study were as follows. First, the measured angle of C-curve is very important to study the evaluation of an ingrowing toenail. Second, it is more fit to study the ingrowing toenail than the self-judged angle. Third, it is differed by the presence of an ingrowing toenail. Four, it is differed among the position of the center of mass during walking. Five, it is narrowed in the case of high hill typed footgear. Six, it is narrowed for the group not taken footcare.

지가변동률 예측을 위한 시계열 모형 분석

우경(Woo Kyoung),이성석(Rhee Sung Suk) 한국부동산학회 2015 不動産學報 Vol.60 No.-

본 연구는 1987년 1/4분기부터 2014년 2/4분기까지 분기별 전국 지가변동률을 이용하여 200만호 주택건설, IMF 외환위기 그리고 미국 서브프라임 모기지 여파를 반영한 개입 ARIMA 모형을 구축하고 이들 세 번의 외부 충격효과가 전국 지가변동률에 미친 영향을 분석하였고, 구축된 모형을 이용하여 2015년도 분기별 전국 지가변동률을 예측하였다. 그 결과 200만호 건설은 1분기의 시차를, IMF 외환위기는 2분기의 시차를, 그리고 미국 서브프라임모기지 위기는 1분기의 시차를 가지고 전국 지가변동률에 영향을 주었음을 확인하였다. 또한 2015년도 분기별 전국 지가변동률을 예측한 결과 1/4분기에는 0.20%, 2/4분기에는 0.15%, 3/4분기에는 0.11% 4/4분기에는 0.08%가 상승할 것으로 예측되었다. 지가에 관한 국내의 연구들은 주로 거시계량경제 모형이나 VAR 모형 등을 이용하여 거시경제지표 중 한 지표로 지가변동률을 포함하여 각 지표 간 상호 인과관계를 규명한 연구가 주를 이루어 왔다. 본 연구는 우리나라의 토지시장에 큰 영향을 미친 세 번의 사건을 고려한 개입 ARIMA 모형을 처음으로 구축하고 분석하였다는데 그 의미가 크다고 할 수 있다. 1. CONTENTS (1) RESEARCH OBJECTIVES The purpose of this study is to analyze the effects of two million houses supply plan in 1989, economic crisis in 1997 and USA subprime mortgage crisis occurred in 2008, and to predict the land price change rate of Korea in 2015. (2) RESEARCH METHOD In this study, a time-series analysis was conducted by using intervention ARIMA model with the quarterly data of land price change rate in Korea from 1987 to 2014. In order to build the intervention ARIMA model as modeling steps, classification, identification and selection of ARIMA model were proceeded. Also with the selected ARIMA model, Intervention ARIMA model was fitted and forecasted for the land price change rate in 2015. SPSS win 12.0 was used for this study. (3) RESEARCH FINDINGS The result of this study can be summarized as follows : the intervention effects of two million houses supply plan occurred after one quarter from the time of the plan announced. The intervention effects of economic crisis occurred after two quarters after Korean Government announced IMF bailout loan are necessary. Finally, the effects of subprime mortgage crisis in USA were occurred after one quarter. 2. RESULTS The results of forecasting quarterly change rate of land price with intervention ARIMA model, the predicted land price change rates are as follows : the land price change rates are predicted as 0.20% in first quarter, 0.15% in the second quarte, 0.11% in the third quarter and 0.08% in fourth quarter. There are few land price studies by using intervention ARIMA in Korea. This study carries an important means from the point of first attempt to forecast land change rate with intervention ARIMA model.

네일 샵의 서비스 품질 요인에 대한 평가

김순심(Soon-Sim Kim),이성석(Sung-Suk Rhee),김현주(Hyeon-Ju Kim) 한국의상디자인학회 2009 한국의상디자인학회지 Vol.11 No.3

상태공간모형에 의한 시계분열분석에 관한 연구

이성석 서원대학교 응용과학연구소 1993 응용과학연구 Vol.2 No.1

This paper considers the concepts and the identification procedures of state space models of stochastic systems. The equivalence between state space model and autoregressive moving average model is discussed. The state vector of state space model can be determined by canonical correlation analysis and then the parameters can be estimated more accurately through maximum likelihood procedure with AIC criterion. The results of computer simulation shows the problem that the large the size of sample is, the more the number of dimensions of state vector is. Espacially when sample size is small, the number of dimensions of state vector determined are strongly downward biased.

군집분석에서 군집갯수에 관한 연구

이성석 서원대학교 응용과학연구소 1997 응용과학연구 Vol.6 No.1

This paper considers the determination of the appropriate number of cluster solutions in some hierarchical clustering procedures. Various measures and statistics to make a decision on the number of clusters are discussed and some of them are calculated using six clustering procedures with a real data set. There seems to be no satisfactory method for determining the number of clusters for any type of cluster analysis. It is, however, suggested that the complete linkage and median clustering procedures are much preferable to detect the true solutions by using the above statistics and measures.