초록

Abstract

Titles have been regarded as having effective clustering features, but they sometimes fail to represent the topic of a document and result in poorly generated document clusters. This study aims to improve the performance of document clustering with titles by suggesting titles in the citation bibliography as a clustering feature. Titles of original literature, titles in the citation bibliography, and an aggregation of both titles were adapted to measure the performance of clustering. Each feature was combined with three hierarchical clustering methods, within group average linkage, complete linkage, and Ward's method in the clustering experiment. The best practice case of this experiment was clustering document with features from both titles by within-groups average method.

초록

구조적 초록은 학술 논문의 주제를 표현하는 역할을 하여 학술 논문을 처리하는데 중요한 요소로 인식되어왔다. 이 연구에서는 구조적 초록을 구성하는 세부 필드의 속성을 4개로 분석하고 초록의 구조를 활용하여 문서 클러스터링에 적용할 수 있는 가능성을 고찰고자 하였다. 구조적 초록의 필드 속성을 문서 클러스터링에 적용한 결과 클러스터링 기법간의 편차가 있었으나 연구 목적이 제공하는 정보량에 비해 주제성이 커서 클러스터링 성능에 가장 큰 영향을 미치고 있는 것으로 나타났다. 또한 분석 결과 특정 필드에 특화되어 출현하는 필드 종속적인 단어가 발생하는 것으로 나타나 필드 종속적인 단어를 배제하고 집단내 평균연결 기법을 적용하였을 때는 클러스터링의 성능이 개선되는 것으로 분석되었다.

Abstract

Structured abstracts have been regarded as an essential information factor to represent topics of journal articles. This study aims to provide an unconventional view to utilize structured abstracts with the analysis on sub fields of a structured abstract in depth. In this study, a structured abstract was segmented into four fields, namely, purpose, design, findings, and values/implications. Each field was compared in the performance analysis of document clustering. In result, the purpose statement of an abstract affected on the performance of journal article clustering more than any other fields. Furthermore, certain types of keywords were identified to be excluded in the document clustering to improve clustering performance, especially by Within group average clustering method. These keywords had stronger relationship to a specific abstract field such as research design than the topic of an article.

초록

This experimental study proposes a multi-document summarization method that produces optimal summaries in which users can find answers to their queries. In order to identify the most effective method for this purpose, the performance of the three summarization methods were compared. The investigated methods are sentence clustering, passage extraction through spreading activation, and clustering-passage extraction hybrid methods. The effectiveness of each summarizing method was evaluated by two criteria used to measure the accuracy and the redundancy of a summary. The passage extraction method using the sequential bnb search algorithm proved to be most effective in summarizing multiple documents with regard to summarization precision. This study proposes the passage extraction method as the optimal multi-document summarization method. 攀＊＊＊ 본 연구는 연세대학교 대학원 박사학위논문의 일부를 요약한 것임.＊＊＊ 연세대학교 문헌정보학과 시간강사(shchoi@lis.yonsei.ac.kr)＊＊＊＊연세대학교 문헌정보학과 교수(ymchung@yonsei.ac.kr) 논문접수일자 : 2004년 8월 27일 게재확정일자 : 2004년 9월 13일攀攀

Abstract

초록

Abstract

The purpose of this study is to identify topic areas of academic library research using two informetric methods; word clustering and Pathfinder network. For the data analysis, 139 articles published in major library and information science journals from 2005 to 2009 were collected from the Korean Science Citation Index database. The keywords that represent research topics were gathered from two sections: an abstract and titles in references. Results showed that reference titles usefully represent topics in detail, and combining abstracts and reference titles can produce an expanded topic map.