This study analyzes the discourse of Korean internet users regarding patient clothing and identifies the changes to structure and content of clothing resulting from infectious disease outbreaks. The analysis draws on texts from Korean blogs, internet cafes, and news articles from 2011 to 2021 related to patient clothing. Using Ucinet 5 and NodeXL 1.0.1 programs, network density, centrality, and cluster analyses were conducted using the Wakita–Tsurumi algorithm. Additionally, Latent Dirichlet Allocation (LDA) topic modeling was applied using Python 3.7 to further explore thematic patterns within the discourse. Throughout the period of study, it was found that users consistently discussed the specific purpose and functionality of patient clothing. Following the outbreak of COVID-19, the distribution and influence of keywords related to the functional aspects of patient clothing, such as “hygiene and safety,” significantly increased. An increased focus was placed on elements such as functionality, activity, autonomy, hygiene, and safety during the pandemic as public health concerns grew. It can be seen that patients increasingly share their experiences online and hospitalization rates surge during health crises; this study provides valuable insights into how the design of patient clothing can be improved through various informatics techniques. It underscores the evolving perception of patient clothing as essential medical equipment during health emergencies. In addition, it offers practical guidance for enhancing designs that better reflect shifting societal concerns, particularly regarding health, safety, and patient comfort.

국내에서 연구된 금융교육 유관 학술논문을 보다 객관적으로 이해하고자 논문 초록에서 추출된 키워드를 중심으로 주요 토픽을 추론하여 포괄적인 담론들을 알아보고자 한다. 연구의 효율성을 높이고 반복될 수 있는 후속과제 연구를 위하여 빅 데이터 분석기법(텍스트 마이닝 - LDA)을 활용하였고 주요토픽에 대한 단어들을 추출하였다. 총 208건의 유관된 학술 논문을 전 처리 한 후에 추출된 명사 32,523건 중 상위빈도 1,201건에 대하여 LDA 토픽모델링을 실시한 결과 16개의 토픽 군이 형성되었다. 최다 빈도의 단어는 “금융이해력” 이었고 다음은 “학생”, “금융소비자” 순이었는데 추론 된 토픽들의 공통적인 주요 요소에는 학교와 학생들에게 교육을 제공하거나 공급하는 과정에 관심을 가지고 있다는 것이었다. 핵심 텍스트와 토픽을 정의하면서 피교육자의 사회적 요구와 성인을 위한 금융교육 관심도가 미흡하여 향후 지속적인 연구영역 확대 가 필요하다는 시사점을 발견하게 되었다.

풍공학분야에 특화된 학술지인 한국풍공학회지에서 발간된 논문에 대해 토픽모델링 기법 중 잠재의미분석(LSA)와 잠재디리 크레할당(LDA)을 적용하여 연구주제 추출의 적합성을 비교 평가하였다. 토픽간의 유사도를 평가하기 위해 문서토픽행렬을 이용한 상 관분석법을 제안하였으며, 이를 적용하여 문서단어행렬의 특성벡터로부터 토픽을 추출하는 LSA 대비 단어의 결합확률을 이용하는 LDA가 토픽 구성단어를 2배 이상 사용하여 보다 독립적인 토픽을 추출하였다는 평가결과를 얻었다. 학술지의 연구주제를 종합하면 ‘building’, ‘bridge’를 ‘연구대상’으로 ‘wind speed’, ‘wind load’, ‘vibration control’을 ‘연구목적’으로 ‘wind tunnel test’, ‘numerical method’의 ‘연구방법’을 사용하였다. 향후 토픽모델링은 연구주제를 ‘연구대상’, ‘연구목적’, ‘연구방법’의 구조적인 결합으로 정의하여 단어의 사용특성을 반영하는 방식으로 개선되어야 할 것으로 사료된다.

This thesis studies two imputation methods, the MCMC method and the EM algorithm, that take care of the problem. The performance of the two methods for the linear (or quadratic) discriminant analysis are evaluated under various types of incomplete observations. Based on simulated experiments, the effect of the imputation using the EM algorithm and the MCMC method are evaluated and compared in terms of the probability of misclassification and the RMSE. This is done for the various cases of incomplete observations. The cases are differentiated by missing rates, sample sizes, and distances between two classification groups. The studies show that the probability of misclassification and the RMSE of the EM algorithm method is lower than the MCMC method. Therefore the imputation using the EM algorithm is more efficient than the MCMC method. And the probability of misclassification of the method that all vectors of observations with missing values are omitted from analysis is lower than the EM algorithm and the MCMC method when the samples size is small and the rate of missing values is extremely big.

In this study, we typified the project for revitalizing the rural area through text analysis using news data, and analyzed the main direction and characteristics of the project. In order to examine the factors emphasized among the issues related to the revitalization of rural areas, we used news data related to 'tourism' and 'livelihood', which are the main keyword of the project to promote rural areas. In the analysis, text mining techniques were used. Topic modeling was conducted on LDA techniques for major projects in 'tourism' and 'livelihood' keyword. Based on this, this study typified the projects that are carried out for the activation of rural areas by topic. As a result of the analysis, it was fount that the topics included in the project were distributed in 11 sub-types(Tourism Promotion, Regional Specialization, Local Festival, Development of Regional Scale, Urban and Rural Exchange, Agricultural Support, Community Forest Management, Improve the Settlement Environment, General Welfare Service, Low Class Support, Others). The characteristics of the rural revitalization projects were examined, and it was confirmed that domestic projects were carried out by tourism-oriented projects. To summarize, the government is making projects to revitalize rural areas through related ministries. Within the structure where the project is spreading to the region, a lot of projects are being carried out. It is understood that the tourism and welfare oriented projects are being carried out in the revitalization project of the domestic rural area. Therefore, in order to achieve the goal of rural revitalization, it is believed that it will be effective to carry out a balanced project to improve the settlement environment of the residents.

The purpose of this study was to suggest the foundation of LDA-based model for infrastructure maintenance cost evaluation. The LDA-based model enables to evaluate not only overall maintenance cost, but also various combination of each cell of risk matrix.

The purpose of this study was to suggest the foundation of LDA-based model for infrastructure maintenance cost evaluation. The LDA-based model enables to evaluate not only overall maintenance cost, but also various combination of each cell of risk matrix.

적어도 2,500년 전에 기원된 바둑은 세상에서 가장 오래된 보드 게임 중의 하나이다. 아직까지 포석 바둑에 대한 이론적 연구는 여전히 미흡하다. 본 연구는 특정 프로기사의 포석을 갖고 훈련용 포석으로부터 얻어낸 클래스로의 인식을 위해 전통적인 선형판별분석 알고리즘을 적용하였다. 상위 10위권 한국 프로기사의 포석을 갖고 클래스-독립 선형판별분석과 클래스-종속 선형판별분석을 수행하였다. 실험 결과 클래스-독립 LDA는 평균 14%의 인식률을, 클래스-종속 LDA는 평균 12%의 인식률을 각각 보였다. 또한 연구 결과 일반적인 상식과 달리 PCA가 LDA보다 더 우월하고, 유클리디언 거리 측정 방식이 결코 LDA보다 뒤지지 않는다는 새로운 사실이 밝혀졌다.