Natural language processing of News (intermediate): rule based modelDaemin Park
NLP of news in news big data analysis systems such as
1) 'NewsSource Beta' (powered by Advanced Institutes of Convergence Technology, Seoul National University)
2) 'Big kinds' (powered by Korea Press Foundation)
News Semantic Network Analysis of Named EntitiesDaemin Park
News Semantic Network Analysis of Named Entities
- named entity recognition: person, organization from news
- tagging topics manually per sentences or articles
- semantic network analysis between persons and organizations
개체명 중심 뉴스 의미 연결망 분석
- 뉴스에서 인명, 기관명을 개체명 인식을 통해 추출
- 주제 태그를 부착
- 사람, 기관, 주제 간의 연결망 분석
- 단어 의미 연결망 한계 극복
Natural language processing of News (intermediate): rule based modelDaemin Park
NLP of news in news big data analysis systems such as
1) 'NewsSource Beta' (powered by Advanced Institutes of Convergence Technology, Seoul National University)
2) 'Big kinds' (powered by Korea Press Foundation)
News Semantic Network Analysis of Named EntitiesDaemin Park
News Semantic Network Analysis of Named Entities
- named entity recognition: person, organization from news
- tagging topics manually per sentences or articles
- semantic network analysis between persons and organizations
개체명 중심 뉴스 의미 연결망 분석
- 뉴스에서 인명, 기관명을 개체명 인식을 통해 추출
- 주제 태그를 부착
- 사람, 기관, 주제 간의 연결망 분석
- 단어 의미 연결망 한계 극복
This paper, first, brings to light some features of social ntworking, introducing the concept of inter-subjectivity, theory of distributed cognition and principle of emergence, also mentioning the concept of information fluency for library communities. Secondly, this paper briefly reviews current library applications of social networking in the world level as well as status in Korea, such as twitter (Micro-blogging/ Presence updates), delicious (Web Resources Sharing), librarything (Cataloguing thru Social Networking: social cataloging web application for storing and sharing personal library catalogs and book lists) and library applications of some mash ups. Widgets, Libraries on FriendFeed and Google Profiles of libraries are also mentioned. Third, open source software platforms are also briefly reviewed in terms of library use. In this, a new paradigm shift of information organization in library field is mentioned: attempts are being made to move from a web of documents to a web of data. Popular Rdf Vocabularies are also briefly introduced. In this, FRBR vocabularies are specially emphasized. Since these are relatively not known to the specialists in other areas. FRBR can easily be implemented as an RDF vocabulary, that could be used to create a universal Linked Data library network. Some library related Linked Data projects are also briefed. Some notions of semantic interoperability are also briefed. Lastly,proposed models for Library apllications of social networking are suggested. Some implications of the use of library applications of social networking are also briefed.
Toward a debating machine: A news sentence network analysis algorithm based o...Daemin Park
This research suggests news sentence network analysis algorithm based on similarity and cooccurence. News contains abundant arguments with facts and quotes those are critical to represent agendas. News sentence network is a semantic network which consists of quotes as nodes. Connectivity is defined by relevance between quotes. Relevance matrix is the sum of similarity matrix calculated by cosine similarity algorithm and cooccurence matrix. This study analyzed 949 quotes from 405 news articles and visualized networks. The results verified that semantic paths were well defined to show the sequence of sub-agendas. News semantic network analysis algorithm can provide a methodology to automatically generate a massive corpus in a sentence level as a training set to develop a debating machine.
| CMS를 활용한 도서관 웹사이트 발전 방향
㈜나인팩토리인터랙티브
02-6009-9149
nine@ninefactory.kr
http://ninefactory.kr/
2014년 10월 1일 국공립대학교 도서관 협의회 학술세미나 발표자료입니다.
- 목차 -
1. 웹의 시대
- 이용자 환경의 변화
- HTML 표준의 변화
- 소셜 웹의 도래
2. CMS(Content Manangement System)
- CMS
- CMS의 개념
- 오픈소스 CMS
3. 도서관 웹사이트 발전 방안
- 웹의 관점
- CMS의 관점
- 서비스의 관점
4. 제안시스템
- 시스템 구성도
5.구현사례
- 부산대학교 도서관
- 해외사례
This paper, first, brings to light some features of social ntworking, introducing the concept of inter-subjectivity, theory of distributed cognition and principle of emergence, also mentioning the concept of information fluency for library communities. Secondly, this paper briefly reviews current library applications of social networking in the world level as well as status in Korea, such as twitter (Micro-blogging/ Presence updates), delicious (Web Resources Sharing), librarything (Cataloguing thru Social Networking: social cataloging web application for storing and sharing personal library catalogs and book lists) and library applications of some mash ups. Widgets, Libraries on FriendFeed and Google Profiles of libraries are also mentioned. Third, open source software platforms are also briefly reviewed in terms of library use. In this, a new paradigm shift of information organization in library field is mentioned: attempts are being made to move from a web of documents to a web of data. Popular Rdf Vocabularies are also briefly introduced. In this, FRBR vocabularies are specially emphasized. Since these are relatively not known to the specialists in other areas. FRBR can easily be implemented as an RDF vocabulary, that could be used to create a universal Linked Data library network. Some library related Linked Data projects are also briefed. Some notions of semantic interoperability are also briefed. Lastly,proposed models for Library apllications of social networking are suggested. Some implications of the use of library applications of social networking are also briefed.
Toward a debating machine: A news sentence network analysis algorithm based o...Daemin Park
This research suggests news sentence network analysis algorithm based on similarity and cooccurence. News contains abundant arguments with facts and quotes those are critical to represent agendas. News sentence network is a semantic network which consists of quotes as nodes. Connectivity is defined by relevance between quotes. Relevance matrix is the sum of similarity matrix calculated by cosine similarity algorithm and cooccurence matrix. This study analyzed 949 quotes from 405 news articles and visualized networks. The results verified that semantic paths were well defined to show the sequence of sub-agendas. News semantic network analysis algorithm can provide a methodology to automatically generate a massive corpus in a sentence level as a training set to develop a debating machine.
| CMS를 활용한 도서관 웹사이트 발전 방향
㈜나인팩토리인터랙티브
02-6009-9149
nine@ninefactory.kr
http://ninefactory.kr/
2014년 10월 1일 국공립대학교 도서관 협의회 학술세미나 발표자료입니다.
- 목차 -
1. 웹의 시대
- 이용자 환경의 변화
- HTML 표준의 변화
- 소셜 웹의 도래
2. CMS(Content Manangement System)
- CMS
- CMS의 개념
- 오픈소스 CMS
3. 도서관 웹사이트 발전 방안
- 웹의 관점
- CMS의 관점
- 서비스의 관점
4. 제안시스템
- 시스템 구성도
5.구현사례
- 부산대학교 도서관
- 해외사례