情报科学 ›› 2021, Vol. 39 ›› Issue (11): 21-29.

• 理论研究 • 上一篇    下一篇

融合主题指纹-引文的方法逻辑:理论阐释与数据呈现

  

  • 出版日期:2021-11-01 发布日期:2021-11-15

  • Online:2021-11-01 Published:2021-11-15

摘要: 【 目的/意义】从文献实体的多特征数据融合的视角构建基于主题指纹-引文耦合的数据融合理论模型,并
对数据融合过程进行实证分析。【方法/过程】对融合主题指纹-引文的方法逻辑进行理论阐释和数学机理分析,以
乳腺小叶癌为案例呈现文献特征项的数据融合过程,通过主题指纹的类别归属与引文期刊所属JCR学科的对比发
现二者在揭示知识表征方面的功能特征。【结果/结论】融合主题指纹-引文的2-模知识网络能够发挥主题指纹和引
文共同揭示学科主题和知识结构的功能;引文信息表征了研究的学科基础、学科背景等稳定性知识结构信息,主题
指纹代表了学科研究前沿、突变主题、新兴趋势等动态性知识主题信息。【创新/局限】融合主题指纹-引文的理论模
型和分析方法是从数据融合层面将内容分析方法与引文分析方法进行结合的有效尝试,未来的研究将聚焦多模知
识网络构建、网络结构分析和量化测度研究,进一步提高该理论模型在知识服务领域应用的科学性、普适性。

Abstract: 【Purpose/significance】From the perspective of multi-feature data fusion of document entities,a theoretical model of data fu?
sion based on subject fingerprint-citation coupling is constructed,and the data fusion process is analyzed empirically.【Method/pro?
cess】First,the theoretical explanation and mathematical mechanism analysis of the method and logic of the fusion of subject finger?
prints and citations are carried out; secondly,breast lobular carcinoma is selected as a case to present the data fusion process of the
subject fingerprint and the citation feature items; finally,the comparison between the category attribution of subject fingerprint and the JCR subject to which the citation journal belongs reveals the functional characteristics in revealing knowledge representation.【Result/conclusion】The 2-mode knowledge network can play the function of subject fingerprints and citations to reveal the subject and knowl? edge structure jointly.The citation information represents the relatively stable knowledge structure information of the subject such as the subject foundation and subject background of the research,and the subject fingerprint represents the dynamic knowledge subject information such as the frontier of subject research,sudden change themes,and emerging trends.【Innovation/limitation】The theoretical model and analysis method of fusing subject fingerprint and citation is an effective attempt to combine content analysis method and ci? tation analysis method from the level of data fusion.Future research will focus on multi-mode knowledge network construction,network structure analysis and quantitative measurement research,so as to further improve the scientificity and universality of the theoretical model in the field of knowledge service.