초록 ▼

■ 과학기술 콘텐츠 큐레이션 체제 기반 연구
ᴼ KISTI 콘텐츠 큐레이션 라이프사이클 모델 개발
ᴼ 인공지능 기술을 적용한 콘텐츠 큐레이션 기술 개발

■ 과학기술 콘텐츠 구축
ᴼ 국내외 학술논문 DB 구축
ᴼ 인물/기관 식별정보 구축 및 연계
ᴼ 중국 과학기술정보 DB 구축

■ 국가 R&D 성과물(보고서 원문, 보고서) 관리 체제 구축
ᴼ 국가 R&D 보고서 원문 등록 및 구축
ᴼ 국가 R&D 논문 성과물 검증 구축
ᴼ 국가 R&D 성과물 콘텐츠 분석을 통한 데이

■ 과학기술 콘텐츠 큐레이션 체제 기반 연구
ᴼ KISTI 콘텐츠 큐레이션 라이프사이클 모델 개발
ᴼ 인공지능 기술을 적용한 콘텐츠 큐레이션 기술 개발

■ 과학기술 콘텐츠 구축
ᴼ 국내외 학술논문 DB 구축
ᴼ 인물/기관 식별정보 구축 및 연계
ᴼ 중국 과학기술정보 DB 구축

■ 국가 R&D 성과물(보고서 원문, 보고서) 관리 체제 구축
ᴼ 국가 R&D 보고서 원문 등록 및 구축
ᴼ 국가 R&D 논문 성과물 검증 구축
ᴼ 국가 R&D 성과물 콘텐츠 분석을 통한 데이터셋 구축

■ 한국 DOI 센터 운영 및 고도화
ᴼ DOI 등록관리 서비스 운영
ᴼ DOI 등록대상 콘텐츠 확대 및 시스템 고도화

(출처 : 초록 6p)

Abstract ▼

Ⅳ. Result of R&D
1. S&T content curation policy and technology development
1) Development of KISTI content curation life cycle model
❍ Suggestions for National and international situations and implications related to content curation
- Investigation of national and international case stu

Ⅳ. Result of R&D
1. S&T content curation policy and technology development
1) Development of KISTI content curation life cycle model
❍ Suggestions for National and international situations and implications related to content curation
- Investigation of national and international case studies on content life cycles
- Drawing the implications from previous studies on content curation and digital curation
❍ Analysis of KISTI content infrastructure environment
- Analysis of characteristics (content, format, etc.) of chief content
- Identification of life cycles according to each content
- Departments, tasks, relationship between departments at each life cycle stage
❍ Development of KISTI content curation life cycle model
- Analysis of life cycles according to each content
- Identification of a content curation life cycle
2) Content curation technology development
❍ Study on Automatic Generation of Scholarly Information Metadata Based on Deep Learning
- Academic document Original PDF based Academic metadata normalization and knowledge methodology In-depth case study
- Semi-automatic construction of domestic academic book collection for automatic generation of academic metadata
- Development and verification of metadata automatic generation engine using RNN, CNN, etc.
❍ Development of Automatic Metadata Classification System based on Semi-Supervised Learning
- Semi-automatic construction of subject classification learning collection that can be mapped to specific classification criteria such as book classification (KDC, DDC) for automatic classification of metadata of academic document Metadata
- Development of Automatic Metadata Classification System based on Learnable Semi-Supervised Learning
❍ Development of Automatic Reference Identification System in Academic Documents
- Identification of research documents in academic journals Pattern base and training collection Semi-automatic construction
- Development and verification of RNN-based reference identification system
❍ Development of a Prototype for Intelligent S&T Database Construction System
- It is possible to utilize the automatic generation and classification of metadata and the automatic identification technology of references for the PDF of the academic documents

2. Domestic & overseas S&T content construction
1) Global scholarly information bibliographic content construction
❍ e-Gate DB construction and supporting application
- Total amount of article metadata in e-Gate DB reached 92 million. e-Gate DB covers domestic and foreign journals metadata and fulltext files.
- Article level metadata : 3,121,475 (Total 92,859,135) / Reference data 453,383,276 (Total 601,861,166)
- Paper journal holding information of academic and research libraries : Total 25,398,862 issues of paper journals from 556 libraries
- Electronic journal licensing information of KESLI consortium member libraries : 1,553,090 journal licensing information from 555 libraries
- Journal title authority data: 2,285 controlled journal titles from 8,071 titles (Total 118,155 controlled journal titles from 118,155 tittles)
- Open Access, Free Access journal information : 3,109,293 (Total 13,024,504)
- Linking information : DOI 9,438,272 (Total 44,672,314) / URL 5,636,760 (Total 72,683,239)
- National licensing overseas journal fulltext files : Total 3,500,329 article fulltext files from 7 global scholarly publishers including Elsevier.
❍ e-Gate DB construction system improvement
- Reference information management module improvement : automatic reference record parsing and identification function development
- Article level identifier lookup function development : DOI, PMID, WOS-AN, SCOPUS, KCI
2) Domestic S&T scholarly information fulltext content construction
❍ Domestic S&T journal information creation
- Creating scholarly article metadata result : 80,086 (Total 1,196,741)
- Registering DOI of scholarly article result : 30,654 (Total 292,292)
- Reference data creation result : 1,600,495 (Total 14,050,912)
❍ Publication of Korea Journal Citation Reports(KJCR) 2017 and Distribution
- Calculation of IF / ZIF / Immediacy Index for KJCR 512 journals
❍ Improvement of KSCD management system and KSCD Quality
- Development of Machine Learning Based Scholarly Article Input System
- New construction data quality level : 99.98% (Article) / 99.99% (Reference)
❍ Metadata collection and creation in collaboration with NRF
- 244 domestic scholarly journal titles out of total 1,134 titles
3) Construction of high-quality disambiguation data
❍ Improvement of disambiguation data quality
- The person disambiguation kernel is developed for the purpose of improving the performance of the person identification data using controlled organization data
- To improve disambiguation data quality, adding more entities in affiliation dictionary, establishment of quality management process for disambiguation data, linking international standard name identifiers
❍ Construction of S&T disambiguation data
- Construction of disambiguation as data 31,842 researchers, 14,811 research institutes and linking heterogeneous S&T content
- Developed a entity disambiguation technology(achieving 94.79% in accuracy) that solves ambiguity of person(author), organization(affiliation information) to give unique identifiers
4) Construction of Chinese-Korean translated S&T content metadata
❍ Translation and construction of around 12 million Chinese scholarly article metadata using automatic translator provided from SYSTRAN(a Korean IT company)
5) KISTI holding information resources development and service
❍ Registration and preservation management of information resources
- Acquisition and Management of journals, books, reports and academic conference materials (Total 71,993 : Books 29,908, Proceedings 17,222, KISTI’s publications 6,279 etc.)
- Procurement of 8 electronic resources for KISTI internal service

3. National R&D outcomes content construction
1) National R&D outcomes registration and construction
❍ Building and improving quality of R&D reports : Registration result 92.74%
❍ Sharing and utilizing the R&D reports : National Library of Korea, National Assembly Library, NDSL & NTIS service, NRF etc.
2) National R&D outcomes papers content construction
❍ Acquisition of OA R&D papers and process improvement : Acquisition result 132,709 papers out of total 407,012 (32.6%)
❍ Improvement of quality and increasement of linking rate R&D papers
❍ Expansion of DB for R&D papers verification
3) Korean scholars paper collection DB creation
4) Promotion activities : Workshops, tutorials, major National R&D project management institutions meeting

4. Korea DOI registration agency management
1) DOI registration service management and technology development
❍ Supporting scholarly society for enhancing the impact of domestic journals
❍ Dissemination of various scholarly society citation style language support
❍ Content type expansion of DOI registration : Patents, thesis & dissertations
❍ Data Literature Interlinking support
2) DOI promotion activities
❍ KISTI deliberated DOI managing agenda as a member of DOI executive committee
❍ Opened tutorials, seminars for DOI registration member institutions

(출처 : SUMMARY 20p)

목차 Contents

• 표지 ... 1
• 제 출 문 ... 4
• 보고서 초록 ... 6
• 요 약 문 ... 8
• SUMMARY ... 17
• CONTENTS ... 26
• 목차 ... 28
• 표목차 ... 30
• 그림목차 ... 32
• 제1장 연구개발과제의 개요 ... 36
• 제1절 목적 및 필요성 ... 36
• 제2절 목표 및 내용 ... 40
• 1. 연구개발과제의 목표 ... 40
• 2. 연구개발과제의 내용 ... 41
• 3. 추진전략 및 방법 ... 44
• 4. 기대효과 ... 46
• 제2장 국내외 기술개발 현황 ... 48
• 제1절 국내현황 ... 48
• 제2절 해외현황 ... 50
• 제3장 연구개발 수행 내용 및 결과 ... 55
• 제1절 과학기술 콘텐츠 큐레이션 정책·기술 연구개발 ... 55
• 1. 콘텐츠 큐레이션 모델 개발 ... 55
• 2. 콘텐츠 큐레이션 기술 개발 ... 58
• 제2절 국내외 핵심 과학기술 콘텐츠 구축 ... 74
• 1. 국가 가용 핵심학술 콘텐츠 구축 ... 74
• 2. 국내 과학기술분야 학술논문 DB구축 ... 86
• 3. KISTI 소장 정보자원 개발 및 서비스 ... 102
• 제3절 국가R&D 성과물 관리 체제 구축 ... 109
• 1. 국가R&D 성과물 개요 ... 109
• 2. 국가R&D 논문 성과물 관리체제 구축 ... 112
• 3. 국가R&D 보고서원문 성과물 관리 체제 구축 ... 124
• 4. 국가R&D 성과물 활용⋅확산을 위한 기술 개발 ... 137
• 제4절 국내 DOI 등록관리 체제 구축 및 운영 ... 147
• 1. 한국 DOI 센터 개요 ... 147
• 2. DOI 홍보 및 협력 활동 ... 149
• 3. DOI 등록관리 및 기술개발 ... 153
• 제4장 목표달성도 및 관련 분야에의 기여도 ... 164
• 제1절 연구개발 목표의 달성도 ... 164
• 1. 과학기술 콘텐츠 큐레이션 정책·기술 연구개발 ... 164
• 2. 국내외 핵심 과학기술 콘텐츠 구축 ... 165
• 3. 국가R&D 성과물 관리체제 구축 ... 167
• 제2절 관련 분야 기여도 ... 173
• 1. 과학기술 콘텐츠 큐레이션 정책·기술 연구개발 ... 173
• 2. 국내외 핵심 과학기술 콘텐츠 구축 ... 174
• 3. 국가R&D 성과물 관리체제 구축 ... 174
• 4. 한국 DOI 센터 운영 및 고도화 ... 175
• 제5장 연구개발결과의 활용계획 ... 176
• 제1절 과학기술 콘텐츠 큐레이션 정책·기술 연구개발 ... 176
• 제2절 국내외 핵심 과학기술 정보 자원 개발 및 구축 ... 176
• 제3절 국가 R&D 성과물(논문, 보고서) 관리 체제 구축 ... 177
• 제4절 한국 DOI 센터 운영 및 고도화 ... 177
• 제6장 참고문헌 ... 178
• 끝페이지 ... 181