MTGCN: A multi-task approach for node classification and link prediction in graph data

作者：

Highlights：

• Correlations between nodes and edges are used to improve task performance.

• Augment information by studying different complex structures in the graph.

• The model jumps out of the local optimal solution through parameter propagation.

摘要

•Correlations between nodes and edges are used to improve task performance.•Augment information by studying different complex structures in the graph.•The model jumps out of the local optimal solution through parameter propagation.

论文关键词：Graph convolutional network,Node classification,Link prediction,Multi-task learning

论文评审过程：Received 2 October 2021, Revised 25 January 2022, Accepted 9 February 2022, Available online 4 March 2022, Version of Record 4 March 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.102902