利用引用信息的关键词提取  

Extraction of Keywords with Citation Information

作　　者：陈翀[1] 罗鹏程[1] 汪十红[1] 

机构地区：[1]北京师范大学

出　　处：《图书情报工作》2014年第1期101-108,116,共9页Library and Information Service

基　　金：教育部科技发展中心“网络时代的科技论文快速共享”专项研究资助课题(项目编号20120240001)研究成果之一

摘　　要：提出一种利用引用信息提取关键词的新方法,将候选词项与引用文献之间的关系抽象为二部图,使用Co-HITS方法迭代计算词项重要性得分至收敛,选出得分最高的词项作为关键词。用ACM数据库中主分类为"信息系统"的论文摘要作为数据集进行评测,结果显示本文所提出的方法优于同类基于图模型计算词项重要度的方法,适用于科学文献和其他具有链接关系的文本集合。在考虑引用信息的情况下,所提取的关键词不但概括原文还能体现原文受到外界关注的内容要点。This paper proposes a new method for keywords extraction with citation information. The relationship between candidate terms and citing papers are abstracted to a bipartite, the import score is computed with the general Co -HITS until convergence, and the top scored terms are selected as the extracted keywords. The paper abstracts dataset classified into ＂information system＂ during 2002 -2011 crawled from ACM digital library is evaluated. The result shows that the method performs better than the state-of-art graph-based method. This method suits for scientific literature and other type of text collection containing rich links. The keywords extracted with it can reflect both the main topics of the original document and the focus outside it.

关 键 词：关键词提取 引用文本 Co-HITS 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]