合作网络中关键研发者潜在重要合作者链接预测  

Research on Link Prediction of Potential Important Partners for Key Developers in the Cooperation Network

作　　者：张玲玲[1] 陈卫静[1] Zhang Lingling;Chen Weijing(Library of University of Electronic Science and Technology of China,􀆯Chengdu Sichuan 611731)

机构地区：[1]电子科技大学图书馆,四川成都611731

出　　处：《情报探索》2022年第3期19-25,共7页Information Research

基　　金：四川省文化和旅游厅2018-2020年度图书情报学与文献学规划项目“科学大数据环境下科研人才网络链接演化研究”(项目编号:WHTTSXM[2018]09)成果。

摘　　要：[目的/意义]重点开展科研人才合作网络中关键研发者的潜在重要合作者链接预测分析,实现精准潜在合作者挖掘预测。[方法/过程]首先,明确关键研发者的研究方向,获取相同研究方向的全球论文数据及作者数据;其次,进行作者数据清洗与消歧,获取包含研究对象在内的最大连通网络;再次,对最大连通网络的所有学者进行研究规模、研究影响力及链接可能性等指标的综合性评估;最后,基于遴选阈值筛选潜在重要合作者,进行效果讨论与反馈,并进行实证分析。[结果/结论]为两位关键研发者预测挖掘出8位、15位符合要求的潜在合作者,经两位关键研发者表明推荐效果较好。[Purpose/significance]The paper focuses on the prediction and analysis of potential important partner links for key developers in the scientific research talent cooperation network,so as to achieve accurate recommendation of potential partners. [Method/process] Firstly,the research direction of key developers is defined,and the global paper data and author data of the same research direction are obtained;Secondly,the author data cleaning and disambiguation are carried out to obtain the largest connected network including the research object;Thirdly,the research scale,research influence and link possibility of all the scholars of the largest connected network are comprehensively evaluated;Finally,potential partners are selected based on the selection threshold,and the effect is discussed and feedback and empirical analysis is made. [Result/conclusion] 8 and 15 potential collaborators are recommended to meet the requirements for two key R & D researchers respectively,and the feedback shows that the recommendation effect is good.

关 键 词：合作者 关键研发者 链接预测 合作网络 

分 类 号：G316[文化科学]