Multi-label graph node classification with label attentive neighborhood convolution

作者：

Highlights：

• A novel neural network-based method for multi-label graph node classification.

• Use one-dimensional convolution modules for node representation learning.

• Use attention mechanism to capture node-label dependencies during training.

• Extensive experiments manifest the superiority of the proposed method.

摘要

•A novel neural network-based method for multi-label graph node classification.•Use one-dimensional convolution modules for node representation learning.•Use attention mechanism to capture node-label dependencies during training.•Extensive experiments manifest the superiority of the proposed method.

论文关键词：Multi-label classification,Graph node classification,Graph convolution,Attention mechanism

论文评审过程：Received 28 November 2020, Revised 16 March 2021, Accepted 16 April 2021, Available online 24 April 2021, Version of Record 13 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115063