Structure information learning for neutral links in signed network embedding

作者：

Highlights：

• We present a novel link prediction method for signed network with neutral links.

• The structure information learning problem for neutral links is solved here.

• Experiments with real-world data are presented.

• NL-SGCN achieves excellent performance in signed network with neutral link.

摘要

•We present a novel link prediction method for signed network with neutral links.•The structure information learning problem for neutral links is solved here.•Experiments with real-world data are presented.•NL-SGCN achieves excellent performance in signed network with neutral link.

论文关键词：Signed network,Link prediction,Neutral links,Graph convolutional network

论文评审过程：Received 21 September 2021, Revised 16 January 2022, Accepted 21 February 2022, Available online 28 February 2022, Version of Record 28 February 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.102917