호흡곤란 환자 퇴원 결정을 위한 벌점 로지스틱 회귀모형

박철용,계묘진,Park, Cheolyong,Kye, Myo Jin 한국데이터정보과학회 2013 한국데이터정보과학회지 Vol.24 No.1

In this paper, penalized binary logistic regression models are employed as statistical models for determining the discharge of 668 patients with a chief complaint of dyspnea based on 11 blood tests results. Specifically, the ridge model based on $L^2$ penalty and the Lasso model based on $L^1$ penalty are considered in this paper. In the comparison of prediction accuracy, our models are compared with the logistic regression models with all 11 explanatory variables and the selected variables by variable selection method. The results show that the prediction accuracy of the ridge logistic regression model is the best among 4 models based on 10-fold cross-validation.

엑셀 VBA를 이용한 이분형 로지스틱 회귀모형 교육도구 개발

박철용,최현석,Park, Cheolyong,Choi, Hyun Seok 한국데이터정보과학회 2014 한국데이터정보과학회지 Vol.25 No.2

Binary logistic regression analysis is a statistical technique that explains binary response variable by quantitative or qualitative explanatory variables. In the binary logistic regression model, the probability that the response variable equals, say 1, one of the binary values is to be explained as a transformation of linear combination of explanatory variables. This is one of big barriers that non-statisticians have to overcome in order to understand the model. In this study, an educational tool is developed that explains the need of the binary logistic regression analysis using Excel VBA. More precisely, this tool explains the problems related to modeling the probability of the response variable equal to 1 as a linear combination of explanatory variables and then shows how these problems can be solved through some transformations of the linear combination.

페널티 방법을 이용한 주성분분석 연구

박철용,Park, Cheolyong 한국데이터정보과학회 2017 한국데이터정보과학회지 Vol.28 No.4

In this study, principal component analysis methods using Lasso penalty are introduced. There are two popular methods that apply Lasso penalty to principal component analysis. The first method is to find an optimal vector of linear combination as the regression coefficient vector of regressing for each principal component on the original data matrix with Lasso penalty (elastic net penalty in general). The second method is to find an optimal vector of linear combination by minimizing the residual matrix obtained from approximating the original matrix by the singular value decomposition with Lasso penalty. In this study, we have reviewed two methods of principal components using Lasso penalty in detail, and shown that these methods have an advantage especially in applying to data sets that have more variables than cases. Also, these methods are compared in an application to a real data set using R program. More specifically, these methods are applied to the crime data in Ahamad (1967), which has more variables than cases.

복합표본 설계를 이용한 건강관련 삶의 질에 영향을 미치는 요인에 관한 연구

박철용,최현석,Park, Cheolyong,Choi, Hyun Seok 한국데이터정보과학회 2014 한국데이터정보과학회지 Vol.25 No.4

Using complex sample design, this article analyzes the difference between sex and age groups of mental health, physical activity, suffering lifestyle diseases, drinking, and smoking using the fifth Korea national health and nutrition examination survey data (2011-2012), and then analyzes the effect of mental health, physical activity, suffering lifestyle diseases, drinking, and smoking on EQ-5D, a measure of health-related quality of life. The results show that mental health, physical activity, suffering lifestyle diseases, drinking and smoking are statistically different among gender and age groups, and that age group, education level, suffering diabetes, recognizing stress, thinking suicide are statistically influential on EQ-5D.

구형 대칭성 검정에 대한 연구

박철용,Park Cheolyong 한국통계학회 2005 응용통계연구 Vol.18 No.1

In this article, we propose a chi-squared test of spherical symmetry. The advantage of this test is that the test statistic and its asymptotic p-value are easy to compute. The limiting distribution of the test statistic is derived under spherical symmetry and its accuracy, in finite samples, is studied via simulation. Also, a simulation study is conducted in which the power of our test is compared with those of other tests for spherical symmetry in various alternative distributions. Finally, an illustrative example of application to a real data is provided.

라소를 이용한 간편한 주성분분석

박철용,Park, Cheolyong 한국데이터정보과학회 2013 한국데이터정보과학회지 Vol.24 No.3

In this study, a simple principal component analysis using Lasso is proposed. This method consists of two steps. The first step is to compute principal components by the principal component analysis. The second step is to regress each principal component on the original data matrix by Lasso regression method. Each of new principal components is computed as the linear combination of original data matrix using the scaled estimated Lasso regression coefficient as the coefficients of the combination. This method leads to easily interpretable principal components with more 0 coefficients by the properties of Lasso regression models. This is because the estimator of the regression of each principal component on the original data matrix is the corresponding eigenvector. This method is applied to real and simulated data sets with the help of an R package for Lasso regression and its usefulness is demonstrated.

스포츠 경쟁력 지수로서 페이지랭크 기반 방법론 사용에 대한 소고

박철용(Cheolyong Park) 한국데이터정보과학회 2021 한국데이터정보과학회지 Vol.32 No.6

최근에 스포츠 경쟁력을 객관적으로 판단하는 방법으로 페이지랭크 (PageRank) 방법이 사용되기 시작하였다. 원래 페이지랭크가 웹사이트의 연결 네트워크 정보를 이용하여 특정 웹사이트의 중요도를 평가하는 것에 착안하여, 특정 운동 경기의 승패 관계를 연결 네트워크로 파악하여 페이지랭크를 해당 운동 경기의 경쟁력 지수로 사용하고자 하는 시도이다. 페이지랭크는 경쟁력이 높은 상대에게 얻은 승리를 경쟁력이 낮은 상대에게 얻은 승리보다 더 높이 평가하는 구조이기 때문에 단순 승패만 따지는 방법보다 경쟁력을 판단하기 더 합리적인 방법으로 여겨진 것이다. 페이지랭크는 연결여부만 고려하기 때문에 가중페이지랭크 (weighted PageRank)로 발전하였고, 학술지 평가에서는 SJR (Scimago Journal Rank)와 아이겐팩터 (Eigenfactor)로 발전하게 되었다. 이 논문에서는 이 방법에서 사용된 공식을 중심으로 각 항목들이 스포츠 경쟁력 관점에서 어떤 의미를 가지는지 살펴본다. 마지막으로 2018년 중학교와 고등학교의 배드민턴 단체 경기 자료에 이 방법들을 적용하고 그 결과를 비교한다. Recently PageRank has started being used as an objective method of measuring a sports competency. This is based on the observation that PageRank measures a website’s importance in a network of websites’ connection and thus can be applied to sports by considering win-loss results of the sports as a network. PageRank can be considered a more reasonable method than simple win-loss counts because PageRank gives more weights to a win over higher rankers than to lower rankers. Since PageRank just considers whether two websites are connected or not, it is evolved into weighted PageRank, and further into SJR (Scimago Journal Rank) and Eigenfactor as journal citation indices. In this article, we will review the formulas of these methods to understand the meaning of each term from the perspective of a sports competency index. Finally we will apply these methods to badminton team match data of middle and high schools in 2018 and then compare the results.

3차원 광메모리를 이용한 인증 기법

박철용(CheolYong Park),류재철(JaeCheol Ryou) 한국정보보호학회 2016 정보보호학회논문지 Vol.26 No.6

최근 사용자 인증에 대한 요구가 증가하면서 패스워드, 그래픽인증, 토큰, 생체인식, 복합 인증 등 다양한 인증 기법이 사용되고 있다. 특히 편의성으로 인해 패스워드, 그래픽, 생체 인증 등 2차원적 요소(2차원 평면)의 데이터를 정보로 이용하고 2차원 모양의 특징 정보를 저장하여 인증에 적용되는데 저장되는 정보는 2차원 정보로 기록 저장된다. 이러한 인증 데이터는 정보를 기록하고 갱신함에 있어 추가적인 데이터 기록을 요구하는 문제가 있다. 또한 2차원 메모리는 저장 데이터의 양이 증가할수록 저장 데이터를 인증 데이터와 비교하는 시간이 증가하여 인증 절차 수행에 소요되는 시간이 증가하는 단점도 있다. 이러한 인증 데이터를 메모리에 저장 및 인증하는 기술에 있어 신속하게 인증 절차를 수행하는 것은 중요한 요소가 되었다. 본 논문에서는 메모리를 사용한 인증 데이터의 기록 및 인증에 있어 기존 방식의 단점을 보완하고자 3차원 광메모리를 이용하여 데이터의 기록시 기록요소를 다양하게 변경하여 서로 다른 저장 데이터를 다중 기록하는 방식을 구성함으로써 고밀도 기록이 가능하게 하고, 데이터의 기록 및 복원시 병렬처리가 가능한 방식을 적용하여 신속한 데이터의 기록 및 복원이 가능한 기법을 제안한다. 또한 인증 데이터 갱신시 기존에 기록된 저장 데이터를 조합하여 조합한 데이터를 인증에 이용함으로써 매번 추가적인 데이터 기록을 요구하는 문제를 해결하고자 한다. 제안한 인증 방식은 3차원 광메모리에 데이터를 기록 저장 후, 복원시 기록 조건에 상응하는 조건을 만족해야만 데이터가 복원되는 방식을 적용한 인증 기법을 제안하고 이를 시험을 통해 인증 기법으로 적용 가능함을 확인하였다. Recently the need for user authentication with increasing, there are a variety of mechanisms, such as password, graphic authentication, token, biometrics and multiple authentication. in particular, the data of the 2-dimensional(2D) factors such as password, graphic authentication, biometrics is used because of the convenience. The stored information is problematic in that additional data recording needs to be performed whenever authentication data is updated. Furthermore, this storage method is problematic in that the time it takes to perform authentication increases because the time it takes to compare storage data with authentication data increases in proportion to an increase in the amount of the storage data. Accordingly, authentication through the rapid comparison of storage data with authentication data is a very important factor in data recording and authentication technology using memory. Using the three-dimensional(3D) optical memory by variously changing the recoding elements during recoding of data constitutes the way that multiple recoding different data storage. This enables high-density recoding in this way, and by applying the possible parallel processing at the time of recording and restoring method, provided that it is possible to quickly record and restore the data. In addition, each time to solve problems that require additional data recorded by a combination of the stored data record in the old data using a combination of the authentication. The proposed mechanism is proposed an authentication method using scheme after the recoding data in 3D optical memory to apply the conditions corresponding to the recoding condition when restoring the recorded data and through the experiment it was confirmed possible application as an authentication mechanism.

광물자원 민족주의의 부상과 EU CRMA 출현에 따른 우리기업의 대응방안

박철용(Cheolyong Park),이하얀(Hayann Lee) 한국정치사회연구소 2023 한국과 국제사회 Vol.7 No.4

토픽 모형 및 사회연결망 분석을 이용한 한국데이터정보과학회지 영문초록 분석

김규하,박철용,Kim, Gyuha,Park, Cheolyong 한국데이터정보과학회 2015 한국데이터정보과학회지 Vol.26 No.1

This article analyzes English abstracts of the articles published in Journal of the Korean Data & Information Science Society using text mining techniques. At first, term-document matrices are formed by various methods and then visualized by social network analysis. LDA (latent Dirichlet allocation) and CTM (correlated topic model) are also employed in order to extract topics from the abstracts. Performances of the topic models are compared via entropy for several numbers of topics and weighting methods to form term-document matrices.