Frog-GNN: Multi-perspective aggregation based graph neural network for few-shot text classification

作者：

Highlights：

• Combining pre-trained language model and GNN for few-shot text classification.

• Extracting both pair and instance-level representations from BERT.

• Capturing intra- and inter-class relationships by multi-perspective aggregation.

• Cumulative loss is adopted to better optimize parameters in multiple GNN layers.

摘要

•Combining pre-trained language model and GNN for few-shot text classification.•Extracting both pair and instance-level representations from BERT.•Capturing intra- and inter-class relationships by multi-perspective aggregation.•Cumulative loss is adopted to better optimize parameters in multiple GNN layers.

论文关键词：Graph neural networks,Multi-perspective aggregation,Few-shot learning

论文评审过程：Received 10 May 2020, Revised 22 January 2021, Accepted 24 February 2021, Available online 5 March 2021, Version of Record 31 March 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114795