• Proceedings of the Korean Statistical Society Conference
• /
• 2004.11a
• /
• pp.277-281
• /
• 2004

최근 마케팅 현업에서 마이크로 마케팅(Micro Marketing)이 마케팅 기법의 화두로 등장하면서 데이터 퓨전(Data Fusion) 또는 데이터 인리치먼트(Data Enrichment)가 각광받는 영역으로 등장하고 있다. 본 연구에서는 데이터 퓨전의 개념과 그를 둘러싸고 있는 통계적 문제와 그 대안에 대하여 논의한다.

• The Journal of Small Business Innovation
• /
• v.20 no.3
• /
• pp.49-68
• /
• 2017

Platform finance is emerging as an alternative finance for SMEs by suggesting a new funding source based on a new technology named FinTech. The essence of this business is the adapting ICT challenges to the financial industry that can adequately reflect risk assessment using Big Data and effectively meet individual risk-return preference. Thus, this is evolving as an alternative to existing finance in the form of P2P loans for Micro Enterprises and supply-chain finance for SMEs that need more working capital. Platform finance in Korea, however, is still at an infant stage and requires policy support. This can be summarized as follows: "Participation of institutional investors and the public sector," meaning that public investors provide seed money for the private investors to crowd in for platform finance. "Negative system in financial regulations," with current regulations to be deferred for new projects, such as Sandbox in the UK. In addition, "Environment for generous use of data," allowing discretionary data sharing for new products," and "Spreading alternative investments," fostering platform finance products as alternative investments in the low interest-rate era.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.89-108
• /
• 2023

This study provides a comprehensive analysis of recent studies conducted on the topic of nowcasting in order to enhance the accuracy and promptness of official statistical data. Furthermore, we propose an alternative approach involving the utilization of real-time data and its corresponding collection methods to effectively operate a real-time nowcasting model capable of accurately capturing the current economic condition. We explore high-frequency real-time data that can predict economic indicators in both the public and private sectors and propose a pipeline for data collection processing and modeling that is based on cloud platforms. Furthermore we validate the essential elements required for the implementation of real-time nowcasting, as well as their data management protocols to ensure the reliability and consistency needed for accurate forecasting of official statistical indicators.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.05a
• /
• pp.1758-1764
• /
• 2006

컨테이너 터미널 간의 치열한 경쟁, 대형 선박의 출현, 컨테이너 터미널 운영의 자동화 등으로 인해서 세계 각국에서는 많은 컨테이너 터미널들을 건설 중이다. 컨테이너 터미널의 건설에는 아주 많은 시간과 비용이 소요되기 때문에 컨테이너 터미널을 효율적으로 설계하는 것이 필수적이다. 본 연구는 컨테이너 터미널의 엔지니어링 프로세스를 지원하기 위해서 개발된 소프트웨어 시스템 TODAY(Terminal Optimization, Design, and Analysis sYstem)에 대해 다루고자 한다. 먼저 설계 프로세스를 제안하였다. 설계 프로세스는 데이터 입력, 능력 소요 계산, 대안 설계, 대안 평가, 결과 출력으로 구성된다. 대안을 신속히 설계하고 평가하기 위해서 컨테이너 터미널 전용 라이브러리가 개발되었다. 라이브러리는 Wharf, Yard, TravellingArea, Gate로 구성된다. 라이브러리의 구조는 객체 지향적으로 설계하여 확장성과 재사용성을 가지게 하였다. 시스템은 사용자에게 친숙한 인터페이스와 CAD 환경을 제공하여, 다양한 유형의 컨테이너 터미널 모형들이 데이터 입력과 라이브러리 구성요소의 조합에 의해 구성될 수 있도록 지원한다. 생성된 대안 모형들은 시뮬레이션과 경제성 모형에 의해 평가된다. 시스템은 생성된 모형을 시뮬레이션 모형으로 자동 변환하는 기능을 제공한다. 피드백 과정을 거쳐서 최종적으로 사용자의 요구사항을 만족시키는 대안이 제시된다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.12
• /
• pp.286-299
• /
• 2017

The Korean performing arts has been facing stagnation due to oversupply, lack of effective distribution system, and insufficient business models. In order to overcome these difficulties, it is necessary to improve the efficiency and accuracy of marketing by using more objective market data, and to secure audience development and loyalty. This study considers the viewpoint that 'Big Data' could provide more general and accurate statistics and could ultimately promote tailoring services for performances. We examine the first case of Big Data analysis conducted by a credit card company as well as Big Data's characteristics, analytical techniques, and the theoretical background of performing arts consumer analysis. The purpose of this study is to identify the meaning and limitations of the analysis case on performing arts by Big Data and to overcome these limitations. As a result of the case study, incompleteness of credit card data for performance buyers, limits of verification of existing theory, low utilization, consumer propensity and limit of analysis of purchase driver were derived. In addition, as a solution to overcome these problems, it is possible to identify genre and performances, and to collect qualitative information, such as prospectors information, that can identify trends and purchase factors.combination with surveys, and purchase motives through mashups with social data. This research is ultimately the starting point of how the study of performing arts consumers should be done in the Big Data era and what changes should be sought. Based on our research results, we expect more concrete qualitative analysis cases for the development of audiences, and continue developing solutions for Big Data analysis and processing that accurately represent the performing arts market.

• Proceedings of the Korean Statistical Society Conference
• /
• 2001.11a
• /
• pp.105-108
• /
• 2001

본 논문에서는 결측치가 있는 불완비 데이터에서 분류나루를 구축하는 방법을 고찰하였다. 기존의 결측치 처리 방법인 대리 분리 방법의 대안으로 대체 방법으로 결측치를 처리한 후 분류나무를 구축하는 방법을 제안하였다.

• Journal of Digital Convergence
• /
• v.13 no.1
• /
• pp.487-493
• /
• 2015

This paper derives the strategic implications of smart learning as a sophisticated alternative to e-learning through the convergence approach of e-learning and Bigdata based on the practices of developed countries. To this, this paper derives e-Learning status and challenges issues in Korea, and then, analyzes the convergence case of e-learning and data science in major foreign advanced companies and universities. In addition, this study conducts an awareness survey on Bigdata applied for employees of e-learning companies, and then derives a strategic alternative to the Bigdata convergence-based smart learning effectiveness in the industry with the analysis of the survey data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.369-370
• /
• 2023

대한민국은 고령화 사회에 접어들었다. 본 연구에서는 서울시 지하철 무임승차를 대표로 선정하여 서울시의 노인인구 특성, 지하철 이용행태를 고려한 정책대안을 모색함으로써 지하철 적자 문제의 해결 방안을 제안하였다. 공공 데이터를 통해 통계분석을 하고, Linear Regression 분석을 통해 결과를 예측하였다. 예측한 결과에 노인의 소득수준을 반영하여 손실 비용이 어떻게 변화하는지 비교하였다.

• Journal of The Korean Association For Science Education
• /
• v.30 no.8
• /
• pp.1110-1122
• /
• 2010

The purpose of this study was to identify elementary school students' alternative conceptual types of the movement of the moon and to find out how these types change when confronted with cognitive conflict situations. To find out alternative conceptual types, 206 sixth graders were sampled, and to investigate how the alternative conceptual types were changed by cognitive conflict situations, and 30 students were systematically resampled by alternative conceptual types. Data were collected through the pre- and post-test instruments, including five items that were used for testing the students' alternative conceptual types and changes after conflict situations. After the pre-test, students were instructed to determine the change of the alternative preconceptions using conflict situations. We found that a majority of students had various kinds of alternative preconceptions formed from their early years of elementary school. The cognitive conflict situations were effective for the conceptual change of the movement of the moon. Specifically, in all groups, the subjects' understanding of "the movement of the early evening crescent moon" changed scientifically.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.432-433
• /
• 2017

최근 ICT 기술과 의료 빅데이터를 활용한 다양한 연구가 활발하게 진행되고 있다. 이질적인 의료데이터의 공유와 확산을 위해 표준화 데이터 모델로 온톨로지 기반의 Linked Open Data가 대안으로 급부상하고 있다. 특히 의료 빅데이터의 분석을 위한 데이터 셋은 프로토콜화하기 어려운 문제점을 갖고 있다. 본 논문에서는 이러한 문제점을 해결하기 개발된 계층적 LOD 개발 환경 시스템을 기반으로 의료정보를 적용하기 위한 모델링에 중점을 두고자 한다. 본 연구는 의료 빅데이터의 검색과 분석연구 분야에 큰 영향을 줄 것으로 기대하고 있다.