arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1601.00288cs.DL
收藏

引文中长期概念符号的识别:文献可以根据共同的知识历史进行聚类吗?

Identification of long-term concept-symbols among citations: Can documents be clustered in terms of common intellectual histories?

Jordan A. Comins, Loet Leydesdorff

摘要:

“引经”不仅被引率高,而且被引时间长达几十年。我们通过比较参考出版年光谱(RPYS)在随后的时间间隔内产生的光谱图中的峰是否表明这种长期影响。多RPYS使我们能够区分在十年内衰减的研究前沿的短期引用峰值与作为概念符号的历史构成(长期)引用(Small,1978)。使用这些构成性引用,能够对文档集进行聚类(例如,在知识共享的历史方面。我们使用多RPYS对Web of Science中的40种信息与图书馆学期刊进行聚类,以验证这一前提。因此,RPYS不仅可以用于检索根的研究(引用),但也为算法的引用集的史学。重要引用是历史上植根于其他引用中的符号,具有货币功能。

英文摘要:

"Citation classics" are not only highly cited, but also cited during several decades. We test whether the peaks in the spectrograms generated by Reference Publication Years Spectroscopy (RPYS) indicate such long-term impact by comparing across RPYS for subsequent time intervals. Multi-RPYS enables us to distinguish between short-term citation peaks at the research front that decay within ten years versus historically constitutive (long-term) citations that function as concept symbols (Small, 1978). Using these constitutive citations, one is able to cluster document sets (e.g., journals) in terms of intellectually shared histories. We test this premise by clustering 40 journals in the Web of Science Category of Information and Library Science using multi-RPYS. It follows that RPYS can not only be used for retrieving roots of sets under study (cited), but also for algorithmic historiography of the citing sets. Significant references are historically rooted symbols among other citations that function as currency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑