摘要
作者引文耦合分析是发现领域活跃研究社群与知识结构的重要工具。当前该方法主要利用引用次数朴素地表征作者间的联系强度,忽视了耦合双方在更深层次上的相似性。为增强现有方法的可靠性与分析深度,本研究试图利用丰富的全文本资源,挖掘引用内容所蕴含的“引了什么”“在何处引”“引用的重要性如何”等关键信息,从施引动机的相似性这一本质层面优化引文耦合强度,提出一种融合引用语义和语境特征的作者引文耦合分析法。该方法通过深入学术论文全文,抽取耦合双方在施引论文中引用内容的语义和语境特征,以此计算增强型引文耦合强度,为每一次引文耦合赋予不同的相似程度值;在此基础上,通过“论文-主题-作者”聚合映射考虑作者的多元兴趣倾向,最终获得作者间的研究主题相似性度量。为证明提出方法的有效性,本研究利用中文“图书情报与数字图书馆”领域的13562篇论文的全文本数据开展了实证研究。实证结果表明,提出方法相较于现有作者引文耦合分析法具有更优的作者兴趣社群发现效果,呈现出更佳、更细致的聚类群落分布,划分出的作者兴趣社群具有更高的群内研究主题同质性和互引概率;此外,提出方法面向大体量作者时的表现更为稳定,具有拓展和应用前景。
Author bibliographic coupling analysis(ABCA) is an important tool for detecting active research communities and mapping domain knowledge structures. ABCA only uses the citation count to naively represent the bibliographic coupling relevance strength between authors, ignoring their similarities at a deep internal level. To enhance the reliability and insight of ABCA, this study attempts to use rich full-text resources to mine key information contained in the citation content, and proposes an innovative method, named semantic-and contextual-based author bibliographic coupling analysis(SC-ABCA), which aims at enhancing the strength of the similarity of citing motives from the essential level. By mining the full-text of scientific literature, the proposed method extracts the features of semantics and context of citations to calculate the enhanced bibliographic coupling strength and assign a different similarity value to each bibliographic couple.Through the“paper-topic-author”aggregation mapping, which considers each author’s interests toward multiple current topics, this method shows the topic similarity relationship between authors. This research also performs an empirical study using 13,562 full-text papers in the field of Chinese LIS. The result shows that the proposed method performs better than ABCA on author community discovery and maps a more detailed knowledge structure. It has a higher content coherence gain and the probability of mutual citation inside clusters;additionally, the method is more robust when facing large-scale author data and has potential for further expansion and application.
作者
张汝昊
袁军鹏
Zhang Ruhao;Yuan Junpeng(Department of Library,Information and Archives Management,School of Economics and Management,University of Chinese Academy of Sciences,Beijing 100049;National Science Library,Chinese Academy of Sciences,Beijing 100190)
出处
《情报学报》
CSSCI
CSCD
北大核心
2022年第8期796-811,共16页
Journal of the China Society for Scientific and Technical Information
基金
国家社会科学基金项目“中外同学科期刊跨遴选体系联合排序方法研究”(19BTQ090)。
关键词
作者引文耦合分析
全文本引文分析
引文内容分析
引用语义
引用语境
领域知识结构
author bibliographic coupling analysis(ABCA)
full-text citation analysis
content-based citation analysis
ci‐tation semantics
citation context
knowledge structure
