强弱连接对学科引证知识扩散动态链路预测的影响研究

doi:10.13266/j.issn.0252-3116.2021.13.007

图书情报工作 ›› 2021, Vol. 65 ›› Issue (13): 66-76.DOI: 10.13266/j.issn.0252-3116.2021.13.007

强弱连接对学科引证知识扩散动态链路预测的影响研究

岳增慧¹, 许海云², 赵敏³

1. 济宁医学院医学信息工程学院日照 276826;
2. 山东理工大学管理学院淄博 255000;
3. 济宁医学院日照 276826

收稿日期:2020-10-10 修回日期:2021-02-20 出版日期:2021-07-05 发布日期:2021-07-10
作者简介:岳增慧(ORCID:0000-0002-9192-9071),副教授,博士,E-mail:yzh66123@126.com;许海云(ORCID:0000-0002-7453-3331),教授,博士;赵敏(ORCID:0000-0002-6190-1481),教授,博士。
基金资助:
本文系国家自然科学基金青年项目"学科知识扩散规律及动力学机制研究"（项目编号：71704063）和国家自然科学基金青年项目"基于科学-技术主题关联分析的创新演化路径识别方法研究"（项目编号：71704170）研究成果之一。

Influence of Strong and Weak Ties on Dynamic Link Prediction of Knowledge Diffusion in Disciplinary Citation Networks

Yue Zenghui¹, Xu Haiyun², Zhao Min³

1. School of Medical Information Engineering, Jining Medical University, Rizhao 276826;
2. Business School, Shandong University of Technology, Zibo 255000;
3. Jining Medical University, Rizhao 276826

Received:2020-10-10 Revised:2021-02-20 Online:2021-07-05 Published:2021-07-10

摘要/Abstract

摘要： [目的/意义] 强弱连接是影响学科引证知识扩散动态链路预测的重要因素之一。学科知识扩散强弱引证连接相互协同、相互影响，共同促进了学科间的知识交流、融合与创新。学科引证知识扩散动态链路预测中强弱连接效应的探索，可为强弱连接理论应用场景的拓展，学科引证知识扩散行为微观演化规律的揭示以及动态链路预测算法指标的评价、设计与优化提供理论与实践参考。[方法/过程] 依托内外协同的思路理念，构建一种外部网络结构调控与内部微观演化机理剖析相结合的动态链路预测强弱连接效应探测方法，分别从学科引证知识关联权重调节、连边失效触发以及强弱连接模体分析三个维度，对基于共同邻居相似性的学科引证知识扩散动态链路预测中的强弱连接效应问题进行探讨。[结果/结论] 强连接在学科引证知识扩散网络演化及动态链路预测过程中扮演着更加重要的角色；链路预测中的强弱连接现象不仅与学科引证关联权重有关，还会受到共同邻居数目以及网络微观模体结构的影响；知识宿学科的吸纳融合能力相对于知识源学科的溢出辐射能力来说，在新连边衍生过程中的主导地位更加突出。

关键词: 学科引证知识扩散, 强弱连接, 权重调节, 连边失效, 三元组模体, 动态链路预测

Abstract: [Purpose/significance] The strong and weak ties is one of the important factors that affect the dynamic link prediction of knowledge diffusion in disciplinary citation networks. The strong and weak citation ties in diffusion of disciplinary knowledge jointly promote knowledge exchange, integration and innovation among disciplines with coordination and mutual effect. The exploration of strong and weak ties in the dynamic link prediction of knowledge diffusion in disciplinary citation networks can provide theoretical and practical references for expanding the application of the strong and weak ties theory, revealing the micro-evolution law of knowledge diffusion behavior of disciplinary citation, and evaluating, designing and optimizing dynamic link prediction algorithm indicators.[Method/process] In this paper, on the basis of synergistic idea, the method of detecting strong and weak ties in the dynamic link prediction was constructed by controlling external structure and analyzing internal evolution mechanism of the networks. To be specific, the influence of strong and weak ties on the dynamic link prediction of knowledge diffusion in disciplinary citation networks based on common neighbor similarity was discussed from three perspectives of adjustment of knowledge connection weight in disciplinary citation networks, link failure triggering and motif analysis.[Result/conclusion] The research has shown that, firstly, strong ties play a more important role in evolution of knowledge diffusion in disciplinary citation networks and dynamic link prediction; secondly, strong and weak ties effect in link prediction is not only related to connection weight of disciplinary citation networks, but affected by the number of common neighbor and micro motif structure; thirdly, compared with the spillover ability of knowledge from source discipline, the absorbing ability of knowledge in destination discipline has a more prominent impact on the process of developing new knowledge links.

Key words: knowledge diffusion in disciplinary citation networks, strong ties and weak ties, weight adjustment, link failure, triad motif, dynamic link prediction

中图分类号:

G250

岳增慧, 许海云, 赵敏. 强弱连接对学科引证知识扩散动态链路预测的影响研究[J]. 图书情报工作, 2021, 65(13): 66-76.

Yue Zenghui, Xu Haiyun, Zhao Min. Influence of Strong and Weak Ties on Dynamic Link Prediction of Knowledge Diffusion in Disciplinary Citation Networks[J]. Library and Information Service, 2021, 65(13): 66-76.

参考文献

[1] GRANOVETTER M S. The strength of weak ties[J]. American journal of sociology, 1973, 78(6):1360-1380.
[2] 吕琳媛,周涛.链路预测[M].北京:高等教育出版社,2013.
[3] 张淑谦,徐顺治,李登科.社会网络理论及其研究述评[J].智库时代,2019(37):264,268.
[4] 滕岩松.基于众包平台LineMe的社会网络演化与透明性研究[D].上海:上海交通大学,2018.
[5] HANSEN M T. The search-transfer problem:the role of weak ties in sharing knowledge across organization subunits[J]. Administrative science quarterly, 1999, 44(1):82-111.
[6] MARTIN R. Strong ties, weak ties and islands:structural and cultural predictors of organizational innovation[J]. Industrial & corporate change, 2002, 11(3):427-449.
[7] FRIEDKIN N E. Information flow through strong and weak ties in intraorganizational social networks[J]. Social networks, 1982, 3(4):273-285.
[8] MILO R, SHEN-ORR S, ITZKOVITZ S, et al. Network motifs:simple building blocks of complex networks[J]. Science, 2002, 298(5594):824-827.
[9] 刘书新.基于非对称神经网络的跨网络社交关系预测研究[D].重庆:重庆邮电大学,2019.
[10] MURATA T, MORIYASU S. Link prediction of social networks based on weighted proximity measure[C]. Proceedings of IEEE/WC/ACM international conference on web intelligence. New York:ACM Press, 2007.
[11] LÜ L Y, ZHOU T. Link prediction in weighted networks:the role of weak ties[J]. EPL (Europhysics letters), 2010, 89(1):18001.
[12] 岳增慧,许海云,王倩飞.基于局部信息相似性的学科引证知识扩散动态链路预测研究[J].情报理论与实践,2020,43(2):84-91,99.
[13] LIU H, HU Z, HADDADI H, et al. Hidden link prediction based on node centrality and weak ties[J]. EPL (Europhysics letters), 2013, 101(1):18004.
[14] SETT N, RANBIR SINGH S, NANDI S. Influence of edge weight on node proximity based link prediction methods:an empirical analysis[J]. Neurocomputing, 2016, 172:71-83.
[15] SHANG K K, SMALL M, XU X K, et al. The role of direct links for link prediction in evolving networks[J]. EPL (Europhysics letters), 2017, 117(2):28002.
[16] LIU B, XU S, LI T, et al. Quantifying the effects of topology and weight for link prediction in weighted complex networks[J]. Entropy, 2018, 20(5):363.
[17] CHEN K J, ZHANG P, YANG Z N, et al. iBridge:inferring bridge links that diffuse information across communities[J]. Knowledge-based systems, 2020, 192:105249.
[18] 吕琳媛.复杂网络链路预测[J].电子科技大学学报,2010,39(5):651-661.
[19] HANLEY J A, MCNEIL B J. The meaning and use of the area under a receiver operating characteristic (ROC) curve[J]. Radiology, 1982, 143(1):29-36.
[20] 岳增慧,方曙.科研合作网络弹性研究与实证[J].图书情报工作,2013,57(11):86-89,95.
[21] 张斌,马费成.科学知识网络中的链路预测研究述评[J].中国图书馆学报,2015,41(217):99-113.
[22] HOLLAND P W, LEINHARDT S. A method for detecting structure in sociometric data[J]. American journal of sociology, 1970, 76(3):492-513.
[23] 刘军.整体网分析:UCINET软件实用指南(第三版)[M].上海:格致出版社,2019.

强弱连接对学科引证知识扩散动态链路预测的影响研究

Influence of Strong and Weak Ties on Dynamic Link Prediction of Knowledge Diffusion in Disciplinary Citation Networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	郑荣, 高志豪, 魏明珠, 王晓宇. 数字中国背景下信息资源管理研究生教育：时代要务、逻辑转向与赋能场景[J]. 图书情报工作, 2023, 67(4): 3-10.
[2]	高旭, 白如江, 王效岳. 面向“卡脖子”技术场景的科技前沿发现与态势演化研究——以集成电路技术为例[J]. 图书情报工作, 2023, 67(4): 40-54.
[3]	欧桂燕, 陈佩, 吴江. 学术文本可读性特征对Altmetrics的影响研究——以Web of Science论文摘要数据为例[J]. 图书情报工作, 2023, 67(4): 102-113.
[4]	朱宇寒, 张均胜, 乔晓东, 张吉玉. 基于科技文献的科技创新供需矩阵构建及评估应用研究[J]. 图书情报工作, 2023, 67(3): 3-15.
[5]	宋凯, 冉从敬. 基于主题挖掘与专利评估的技术机会识别研究——以智慧农业为例[J]. 图书情报工作, 2023, 67(3): 61-71.
[6]	李树青, 黄金旺, 马丹丹, 张志旺. 基于显隐式信息融合和单类协同过滤方法的主题词推荐[J]. 图书情报工作, 2023, 67(3): 72-84.
[7]	王丹丹, 任婧媛. 国外主要社会科学数据管理平台建设研究及启示[J]. 图书情报工作, 2023, 67(3): 131-139.
[8]	邓胜利, 许家辉, 夏苏迪. 数字环境下大学生算法素养评价体系及实证研究[J]. 图书情报工作, 2023, 67(2): 23-32.
[9]	张庭玮, 朱沁雨, 宋德凤, 陈烨. 过滤气泡情境下用户行为及其形成机制研究[J]. 图书情报工作, 2023, 67(2): 64-75.
[10]	梁少博, 尉子仪, 臧岚. 移动搜索会话中的后续行为研究[J]. 图书情报工作, 2023, 67(2): 76-85.
[11]	冯凌子, 张汝昊, 冯凯悦, 袁军鹏. 基于科研产出关联的大科学装置集群规划布局分析方法研究——以粤港澳大湾区综合性国家科学中心为例[J]. 图书情报工作, 2023, 67(2): 86-97.
[12]	肖纪文. 面向局部可解释性机器学习的数据故事生成方法研究[J]. 图书情报工作, 2023, 67(2): 98-107.
[13]	任超, 杨孟辉, 杨冠灿, 霍朝光, 卢小宾. 基于知识图谱的循证政策中科学证据推荐研究——以新冠肺炎疫情防控政策为例[J]. 图书情报工作, 2023, 67(2): 108-118.
[14]	曾江峰, 刘园园, 程征, 段尧清. 基于UniLM模型的学术文摘观点自动生成研究[J]. 图书情报工作, 2023, 67(2): 131-139.
[15]	刘细文. 贯彻落实二十大精神，开创文献情报工作的高质量发展道路[J]. 图书情报工作, 2023, 67(1): 4-8.