引文耦合网络分析在论文评价中的应用  

Application of Bibliographic Coupling Network Analysis in Paper Evaluation

作　　者：于涵 杨颖[2] 崔雷[3] YU Han;YANG Ying;CUI Lei

机构地区：[1]辽宁中医药大学图书馆,辽宁沈阳110847 [2]中国医科大学图书馆,辽宁沈阳110122 [3]中国医科大学健康管理学院,辽宁沈阳110122

出　　处：《图书情报导刊》2024年第12期54-61,共8页Journal of Library and Information Science

摘　　要：探索基于引文耦合的论文相似性网络是否可以用于识别某一领域的重要论文。基于论文的引文耦合相似性构建网络,分析FacultyOpin‐ions推荐论文与非FacultyOpinions推荐论文在网络属性上的差异是否可用于论文评价。结果显示,特定领域重要论文组成的网络密度较大,平均路径长度、连通分量数目较小,孤立点极少,具有小世界属性,表明特定领域内重要论文研究主题集中,论文之间知识传播、信息交换和合作研究比较多。特定领域重要论文与非重要论文节点在度、离心率、接近中心性、中介中心性、PageRank、三角形数量和特征向量中心度方面有显著差异。研究发现,特定领域重要论文与非重要论文在引文耦合相似性网络中网络属性具有差异性,使用网络属性评价论文具有一定意义。Explore whether the paper similarity network based on citation coupling can be used to identify important papers in a certain field.Based on the citation coupling similarity of papers,a network is constructed to analyze whether the differences in network attributes between Faculty Opinions recommended papers and non-Faculty Opinions recom‐mended papers can be used for paper evaluation.The results show that the network density of important papers in specific fields is relatively high,with a small average path length and a small number of connected components,and few isolated points,indicating a small-world property.This suggests that the research topics of important papers in specific fields are concentrated,and there is a lot of knowledge dissemination,information exchange,and collaborative research between papers.There are significant differences in degree,eccentricity,closeness centrality,betweenness centrality,PageRank,number of triangles,and eigenvector centrality between important and non-important papers in specific fields.The study reveals that there are differences in network attributes between important and non-important papers in specific fields in the citation coupling similarity network,and using network attributes to evaluate papers has certain significance.

关 键 词：论文评价 引文耦合 社会网络分析 FacultyOpinion 

分 类 号：G252.8[文化科学—图书馆学]