图书情报工作

图书情报工作 >

2025 , Vol. 69 >Issue 7: 42 - 52

DOI: https://doi.org/10.13266/j.issn.0252-3116.2025.07.004

研究论文

基于早期知识结构新颖性的学者影响力预测研究——以生物医学领域为例

  • 吴志祥 ,
  • 伊海港 ,
  • 祝艺萌 ,
  • 王培 ,
  • 王昊
展开
  • 1 南京工业大学经济与管理学院 南京 211816;
    2 南京大学信息管理学院 南京 211023
吴志祥，副教授，博士，硕士生导师，E-mail:cnwzx2012@njtech.edu.cn;伊海港，硕士研究生;祝艺萌，硕士研究生;王培，硕士研究生;王昊，教授，博士，博士生导师。

收稿日期: 2024-06-25

  修回日期: 2024-09-21

  网络出版日期: 2025-03-24

基金资助

本文系国家自然科学基金项目“基于深度学习与语义关联的关键核心技术—专家组合模型研究”（项目编号：7190408）和国家社会科学基金“功勋科学家文献资源的知识关联与价值挖掘研究”（项目编号：23CTQ027）研究成果之一。

收起

Predicting Scholar’s Influence Based on the Novelty Indicators of Early Knowledge Structure: A Case Study in the Biomedical Field

  • Wu Zhixiang ,
  • Yi Haigang ,
  • Zhu Yimeng ,
  • Wang Pei ,
  • Wang Hao
Expand
  • 1 School of Economics and Management, Nanjing Tech University, Nanjing 211816;
    2 School of Information Management, Nanjing University, Nanjing 211023
Wu Zhixiang,associate professor,PhD,master supervisor,E-mail:cnwzx2012@njtech.edu.cn;Yi Haigang,master candidate;Zhu Yimeng,master candidate;Wang Pei,master candidate;Wang Hao,professor,PhD,doctoral supervisor.

Received date: 2024-06-25

  Revised date: 2024-09-21

  Online published: 2025-03-24

Supported by

This work is supported by the National Natural Science Foundation of China project titled “Research on Key Core Technologies-Experts Combination Model Based on Deep Learning and Semantic Association” (Grant No. 7190408) and the National Social Science Fund of China project titled “Research on Knowledge Association and Value Mining of Documentary Resources of Chinese Meritorious Scientists” (Grant No. 23CTQ027).

Fold

摘要

[目的/意义] 基于学者知识结构蕴含的创新特征，设计学者早期知识结构新颖性的测度指标，以此预测学者未来的影响力，为学术人才的早期识别提供新的指标借鉴。[方法/过程] 首先，通过PubMed Knowledge Graph数据库获取57 927位生物医学领域学者数据，利用受控主题词共现关系构建学者的知识结构；其次，从知识主题与结构位置两个层面出发设计6项指标测度学者早期知识结构新颖性；之后，根据学者后期的影响力对学者进行分类标注，训练机器学习模型；最后，实验评估不同组合变量模型下的分类效果，分析基于知识结构新颖性指标的预测性能。[结果/结论] 新颖性指标能有效预测影响力；单指标预测中，主题新颖性（TN）效果最好，结构层面的4个指标效果均超过主题组合新颖性（TCN）；综合指标的预测F1值平均提高2.7%。从内容特征角度出发，为预测与理解学者学术影响力提供新视角，所设计的新指标具有实用价值，能帮助弥补现有预测指标的不足。

关键词： 影响力预测; 学者知识结构; 新颖性; 机器学习

本文引用格式

吴志祥 , 伊海港 , 祝艺萌 , 王培 , 王昊 . 基于早期知识结构新颖性的学者影响力预测研究——以生物医学领域为例[J]. 图书情报工作, 2025 , 69(7) : 42 -52 . DOI: 10.13266/j.issn.0252-3116.2025.07.004

Abstract

[Purpose/Significance] This study designs indicators to measure the novelty of a scholar’s early knowledge structure based on the innovative characteristics inherent in it. The purpose is to predict the scholar’s future influence, thereby providing new indicators for the early identification of academic talents. [Method/Process] Firstly, a dataset of 59,207 scholars in the biomedical field was obtained from the PubMed knowledge graph (PKG) database, and the knowledge structure of scholars was constructed using the co-occurrence relationships of controlled subject terms. Secondly, 6 indicators were designed to measure the novelty of scholars’ early knowledge structures on two levels: knowledge units and structural positions. Then, scholars were classified and labeled based on their later influence, and some machine learning models were trained. Finally, the classification effectiveness of different models was evaluated using the combination variable method, and the predictive performance of indicators based on the novelty of knowledge structures was analyzed. [Result/Conclusion] The study finds that novelty indicators can effectively predict influence. Among single-indicator predictions, topic novelty (TN) performs the best, while the 4 structural indicators all outperform the topic combination novelty (TCN). The F1 value of the comprehensive indicators is improved by an average of 2.7%. The novelty of knowledge structures provides a perspective for predicting and understanding scholars’ academic influence from content characteristics. The indicators designed are valuable and can complement existing prediction indicators.

Key words： influence prediction; knowledge structure of scholar; novelty indicators; machine learning

参考文献

[1] XIA W, LI T, LI C. A review of scientific impact prediction: tasks, features and methods[J]. Scientometrics, 2023, 128(1): 543-585.
[2] MUSTAFA G, RAUF A, AFZAL M T. GK index: bridging Gf and K indices for comprehensive author evaluation[J]. Knowledge and information systems, 2024, 66: 5203-5238.
[3] 刘萍, 石湘, 邢玉艳. 卓越学者隐性知识体系建模与表示方法 [J]. 情报学报, 2023, 42(8): 915-925. (LIU P, SHI X, XING Y Y. Modeling and representation of tacit knowledge infrastructure of outstanding scholars[J]. Journal of the China Society for Scientific and Technical Information, 2023, 42(8): 915-925.)
[4] ZENG A, SHEN Z, ZHOU J, et al. Increasing trend of scientists to switch between topics[J]. Nature communications, 2019, 10(1): 1-11.
[5] YU H, MARSCHKE G, ROSS M B, et al. Publish or perish: selective attrition as a unifying explanation for patterns in innovation over the career[J]. Journal of human resources, 2023, 58(4): 1307-1346.
[6] HUANG S, LU W, BU Y, et al. Revisiting the explorationexploitation behavior of scholars’ research topic selection: Evidence from a large-scale bibliographic database[J]. Information processing & management, 2022, 59(6): 103110.
[7] PARK M, LEAHEY E, FUNK R J. Papers and patents are becoming less disruptive over time[J]. Nature, 2023, 613(7942): 138-144.
[8] 梁兴堃. 守正能否创新?—— 基于我国图情领域论文新颖性和传统性的分析[J]. 图书情报工作, 2022, 66(20): 148-161. (LIANG X K. Novelty, conventionality, and scientific impact of papers in library and information science in China: evidence from papers in CSSCI (2000-2019) [J]. Library and information service, 2022, 66(20): 148-161.)
[9] 秦成磊, 章成志. 大数据环境下同行评议面临的问题与对策[J]. 情报理论与实践, 2021, 44(4): 99-112. (QIN C L, ZHANG C Z. Problems and responses strategy of peer review in the context of big data[J]. Information studies: theory & application, 2021, 44(4): 99-112.)
[10] XU H, LUO R, WINNINK J, et al. A methodology for identifying breakthrough topics using structural entropy[J]. Information processing & management, 2022, 59(2): 102862.
[11] MIN C, BU Y, SUN J. Predicting scientific breakthroughs based on knowledge structure variations[J]. Technological forecasting and social change, 2021, 164: 120502.
[12] ZHANG T, TAN F, YU C, et al. Understanding relationship between topic selection and academic performance of scientific teams based on entity popularity trend[J]. Aslib journal of information management, 2023, 75(3): 561-588.
[13] JIA T, WANG D, SZYMANSKI B K. Quantifying patterns of research-interest evolution[J]. Nature human behavior, 2017, 1(4): 0078.
[14] 滕广青, 贺德方, 彭洁, 等. 结构与秩序: 知识组织领域中结构主义思想的演进[J]. 情报理论与实践, 2015, 38(4): 6-10. (TENG G Q, HE D F, PENG J, et al. Structure and order: the evolution of thought on structuralism in knowledgeorganization[J]. Information studies: theory & application, 2015, 38(4): 6-10.)
[15] BROOKES B C. The foundations of information science. Part I. Philosophical aspects[J]. Journal of information science, 1980, 2(3/4): 125-133.
[16] 吴志祥, 何超, 赵凯蕊, 等. 基于主题词分布测度的学者知识累积行为研究[J]. 情报理论与实践, 2022, 45(7): 140-147. (WU Z X, HE C, ZHAO K R, et al. Research on the knowledge accumulation behavior of scholars based on the measurement of subject headings distribution[J]. Information studies: theory & application, 2022, 45(7): 140-147.)
[17] 仇鹏飞, 孙建军, 闵超. 科学研究中的师承关系评述与思考[J]. 图书与情报, 2018(5): 50-55, 118. (QIU P F, SUN J J, MIN C. Review of and thoughts on mentor-ship in scientific research[J]. Library and information, 2018(5): 50-55, 118.)
[18] 陈果, 赵以昕. 多因素驱动下的领域知识网络演化模型: 跟风, 守旧与创新[J]. 情报学报, 2020, 39(1): 1-11. (CHEN G, ZHAO Y X. A network evolution model for domain knowledge driven by multiple factors: Following suit, conservatism, and innovation[J]. Journal of the China Society for Scientific and Technical Information, 2020, 39(1): 1-11.)
[19] 周朝阳, 贺艳菊, 柳路芳. 融合合著者多样性及影响力的学术新星预测方法[J]. 情报理论与实践, 2020, 43(2): 78-83, 71. (ZHOU C Y, HE Y J, LIU L F. A method for predicting rising stars by combining co-authors’ diversity and influence[J]. Information studies: theory & application, 2020, 43(2): 78-83, 71.)
[20] HIRSCH J E. An index to quantify an individual’s scientific research output[J]. Proceedings of the national academy of sciences, 2005, 102(46): 16569-16572.
[21] ACUNA D E, ALLESINA S, KORDING K P. Predicting scientific success[J]. Nature, 2012, 489(7415): 201-202.
[22] LINDAHL J. Predicting research excellence at the individual level: The importance of publication rate, top journal publications, and top 10% publications in the case of early career mathematicians[J]. Journal of informetrics, 2018, 12(2): 518-533.
[23] PANAGOPOULOS G, TSATSARONIS G, VARLAMIS I. Detecting rising stars in dynamic collaborative networks[J]. Journal of informetrics, 2017, 11(1): 198-222.
[24] LINDAHL J, COLLIANDER C, DANELL R. Early career performance and its correlation with gender and publication output during doctoral education[J]. Scientometrics, 2020, 122(1): 309-330.
[25] NIE Y, ZHU Y, LIN Q, et al. Academic rising star prediction via scholar’s evaluation model and machine learning techniques[J]. Scientometrics, 2019, 120(2): 461-476.
[26] DAUD A, SONG M, HAYAT M K, et al. Finding rising stars in bibliometric networks[J]. Scientometrics, 2020, 124: 633-661.
[27] 董克, 陈晓萍, 吴佳纯. 科研论文创新性与引文影响力相关性研究—— 基于语义视角的测度[J]. 情报理论与实践, 2023, 46(10): 24-31. (DONG K, CHEN X P, WU J C. Research on the correlation between creativity and citation impact of scientific research paper: measurement from semantic perspective[J]. Information studies: theory & application, 2023, 46(10): 24-31.)
[28] YAN Y, TIAN S, ZHANG J. The impact of a paper ’s new combinations and new components on its citation[J]. Scientometrics, 2020(122): 895-913.
[29] RUAN X, AO W, LYU D, et al. Effect of the topic-combination novelty on the disruption and impact of scientific articles: evidence from PubMed[J]. Journal of information science, 2023: 1-14.
[30] 何婕君, 闵超. 引文持续增长型论文特征和影响因素研究[J]. 图书情报工作, 2022, 66(7): 110-119. (HE J J, MIN C. Research on the features and influencing factors of papers with continuous citation growth[J]. Library and information service, 2022, 66(7): 110-119.)
[31] BORNMANN L. How are excellent (highly cited) papers defined in bibliometrics? A quantitative analysis of the literature[J]. Research evaluation, 2014, 23(2): 166-173.
[32] TU Y N, SENG J L. Indices of novelty for emerging topic detection[J]. Information processing & management, 2012, 48(2): 303-325.
[33] 陆泉, 秦雨萱, 陈静. 跨学科“ 技术— 主题” 创新组合识别—— 以人工智能技术驱动图情领域创新为例[J]. 图书 情报 工作, 2024, 68(2): 50-61. (LU Q, QIN Y Y, CHEN J. Identification of interdisciplinary “technology-topic” innovation combinations: take artificial intelligence technology driving the innovation in LIS as an Example[J]. Library and information service, 2024, 68(2): 50-61.)
[34] XU J, KIM S, SONG M, et al. Building a PubMed knowledge graph[J]. Scientific data, 2020, 7(1): 205.
Options
文章导航

/

〈 〉