• The Journal of Society for e-Business Studies
• /
• v.27 no.2
• /
• pp.1-20
• /
• 2022

As the global business environment is rapidly changing due to the 4th industrial revolution, new jobs that did not exist before are emerging. Among them, the job that companies are most interested in is 'Data Scientist'. As information and communication technologies take up most of our lives, data on not only online activities but also offline activities are stored in computers every hour to generate big data. Companies put a lot of effort into discovering new opportunities from such big data. The new job that emerged along with the efforts of these companies is data scientist. The demand for data scientist, a promising job that leads the big data era, is constantly increasing, but its supply is not still enough. Although data analysis technologies and tools that anyone can easily use are introduced, companies still have great difficulty in finding proper experts. One of the main reasons that makes the data scientist's shortage problem serious is the lack of understanding of the data scientist's job. Therefore, in this study, we explore the job competencies of a data scientist by qualitatively analyzing the actual job posting information of the company. This study finds that data scientists need not only the technical and system skills required of software engineers and system analysts in the past, but also business-related and interpersonal skills required of business consultants and project managers. The results of this study are expected to provide basic guidelines to people who are interested in the data scientist profession and to companies that want to hire data scientists.

• Proceedings of the Korea Contents Association Conference
• /
• 2017.05a
• /
• pp.71-72
• /
• 2017

최근 빅데이터 분석력 강화와 과학기술정보 콘텐츠 기반의 R&D 전략 수립의 중요성이 강조됨에 따라 국내외 주요 과학기술 통계와 정보를 수집, 범주화하여 DB를 구축하고 국가 기초연구지원을 위한 정책을 수립하는데 활용하고자 하였다. 과학기술 통계 정보 콘텐츠 DB를 기반으로 빅데이터 심층분석을 실시하여 우리나라 R&D 경쟁력과 효과성을 보여주고 본 DB 활용성의 확장에 대해 고찰하였다.

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.211-212
• /
• 2011

스트림 프로세싱 시스템은 실시간 데이터 수집 장치와 대규모 분산 컴퓨팅 환경이 결합되어 데이터 생성과 가공을 통하여 다수의 결과를 병렬적으로 도출하는 분산 프로그래밍 모델이다. 본 논문에서는 프로세스간에 필수적으로 요구되는 유입데이터 버퍼 관리에 초점을 두고 있다. 데이터의 유입률에 따라 동적으로 분석 프로세스를 확장시킴으로서 프로세스간 버퍼의 크기를 제어하는 기법을 제안하며 시뮬레이션을 통하여 성능 분석을 하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.18-20
• /
• 2020

본 연구에서는 고차원 데이터에 대한 차원축소 및 군집 분석과 같은 비지도 학습 알고리즘에 대해 알아보기 위해서 얼굴 이미지 데이터 셋을 사용한다. 얼굴 데이터 셋에 대하여 주요 비지도 학습 알고리즘을 이용하여 실시간으로 클러스터링하고, 그 성능을 비교한다. 비디오에서 추출된 영상 속의 7명의 인물에 대하여 Scikit-learning 라이브러리에서 제공하는 클러스터링 알고리즘과 더불어 주요 차원축소 알고리즘(Dimension Reduction Algorithm)을 사용하여 총 10개의 알고리즘에 대하여 분석한다. 또한, 클러스터링 성능 검사를 통해 알고리즘의 성능을 비교해보고, 이를 통하여 앞으로의 연구 방향에 대해 고찰한다.

• Annual Conference of KIPS
• /
• 2024.05a
• /
• pp.461-463
• /
• 2024

최근 ESG 경영 등 환경에 대한 관심이 고조됨에 따라, 기존 화학산업을 대체할 수 있는 바이오화학산업이 성장하고 있다. 바이오화학산업규모는 연평균 성장률 10%로 2050년에는 화학산업 시장의 약 50% 정도를 차지할 것으로 예상될 정도로 유망 분야로 성장하고 있다. 본 논문에서는 신산업으로 성장하고 있는 바이오화학분야의 연구자들이 해당 분야의 유망 소재에 대하여 최신 연구정보를 빠르게 파악하고, 미래 유망 바이오화학물질의 발굴등 바이오화학 분야에 다양하게 활용할 수 있도록 관련 논문 정보를 수집, 저장, 검색할 수 있는 시스템을 개발하였다. 해당 수집 논문정보는 바이오화학산업분류와 연관된 바이오화학물질에 대한 정보와 연계되어 있어, 향후 인공지능 데이터 분석 등에 활용할 수 있는 데이터를 제공할 수 있을 것이라 기대한다.

• The Journal of the Korea Contents Association
• /
• v.21 no.5
• /
• pp.47-69
• /
• 2021

Data scientists are new knowledge workers representing the knowledge economy era. Knowledge workers perform unstandardized works that solve ambiguity-intensive problems. Therefore, self-leadership, which emphasizes self-motivated, autonomous judgment and execution, significantly influences their work-related outcomes. Even knowledge workers have high occupational commitment, they usually show low organizational commitment. Knowledge workers' intention to leave is also relatively high due to this reason. This study focused on data scientists' self-leadership, predicted that self-leadership would increase an organization's commitment and intention to leave. Based on the trait activation theory(TAT), the author also confirmed how perceived job autonomy enhances self-leadership influences. Results showed that data scientists' self-leadership significantly lowered intention to leave through organizational commitment and this mediating effect was moderated by perceived job autonomy. This study broadened the theoretical understanding the effects of knowledge workers' self-leadership and presented practical implications for managing data scientists.

• The Journal of the Korea Contents Association
• /
• v.17 no.12
• /
• pp.537-544
• /
• 2017

Science and technology information data is steadily increasing not only in Korea but also around the world. At present, dozens of documents are produced and distributed every day regardless of field, and researchers are increasingly demanding to use this information effectively to support research activities that are valuable and useful for human life. In the past, researchers are moving away from searching for necessary information, and now they are changing to services that connect data and data, find high value-added information that can be used, and present it to researchers. In this study, we analyze literature data created by Korean researchers from domestic and foreign scientific and technological information data, find use cases, and provide information to help researchers.

• Convergence Security Journal
• /
• v.21 no.2
• /
• pp.3-10
• /
• 2021

Nowadays, data-based scientific research is a trend, and the transmission of large amounts of data has a great influence on research productivity. To solve this problem, a separate network structure for transmitting large-scale scientific big data is required. ScienceDMZ is a network structure designed to transmit such scientific big data. In such a network configuration, it is essential to establish an access control list(ACL) for users and resources. In this paper, we describe the R&E Together project and the network structure implemented in the actual ScienceDMZ network structure, and define users and services to which access control policies are applied for safe data transmission and service provision. In addition, it presents a method for the network administrator to apply the access control policy to all network resources and users collectively, and through this, it was possible to achieve automation of the application of the access control policy.

• Journal of Scientific & Technological Knowledge Infrastructure
• /
• s.8
• /
• pp.113-118
• /
• 2002

한국과학기술정보연구원 (KISTI) 슈퍼컴퓨팅센터의 SeeMore는 국내에서 처음으로 구현된 시스템으로 지난 해 4월에 설치되었으며, 복잡하고 방대한 계산결과를 가상현실 공간상에서 시각화하여, 효과적인 해석을 가능하게 하는 과학적 가시화 시스템이라고 할 수 있다. SeeMore에서 활용될 수 있는 적용 분야는 가상현실, 과학적 데이터의 시각화, 복잡한 데이터의 분석, 기상모델링, 양자화학, 생물정보학, 물리적 현상의 모델링에 이르기까지 무궁무진하다.

• Journal of Korean Library and Information Science Society
• /
• v.36 no.2
• /
• pp.283-305
• /
• 2005

Data produced during scientific and technical activities usually have value as standard reference data (SRD) which is a well-documented numeric value assessed for reliability and accuracy. The main criteria in the evaluation for SRD is how well their production is documented including a sample preparation, measurement method, data processing and so on, other than reference data. SRD is represented as a certified numeric value with uncertainty. In this study, the SRD in the area of material properties was introduced to provide understanding of SRD and its evaluation method. Also the national SRD system was studied. The national SRD system is composed of the center for the SRD, the data centers, and the committee. The role and task of those components were studied. The legislational and systematic supports for the system were proposed in this study.