Generalization of bibliographic coupling and co-citation using the node split network
作者：
Highlights：
• We propose two novel approaches to emulate and generalize citation-based coupling measures using the node split method, which simply replicates a node into two separated nodes: a citing node and a cited node.
• Personalized PageRank on the node split network can imitate bibliographic coupling (BC, from citing nodes) and co-citation (CC, cited nodes) similarities, yet neural embedding does not yield high accuracy.
• The Personalized PageRank networks can also be used as generalized similarity networks accounting from the perspectives of BC and CC.
• We find that many strong relations obtained from the generalized networks are omitted in the coupling networks.
• Comparative tests on global and local sampling strategies suggest that local sampling is more stable for Personalized PageRank.
摘要
•We propose two novel approaches to emulate and generalize citation-based coupling measures using the node split method, which simply replicates a node into two separated nodes: a citing node and a cited node.•Personalized PageRank on the node split network can imitate bibliographic coupling (BC, from citing nodes) and co-citation (CC, cited nodes) similarities, yet neural embedding does not yield high accuracy.•The Personalized PageRank networks can also be used as generalized similarity networks accounting from the perspectives of BC and CC.•We find that many strong relations obtained from the generalized networks are omitted in the coupling networks.•Comparative tests on global and local sampling strategies suggest that local sampling is more stable for Personalized PageRank.
论文关键词：Node split network,Bibliographic coupling,Co-citation,Neural embedding,Personalized PageRank
论文评审过程：Received 28 October 2021, Revised 12 April 2022, Accepted 14 April 2022, Available online 26 April 2022, Version of Record 26 April 2022.
论文官网地址：https://doi.org/10.1016/j.joi.2022.101291
