BertGCN: Transductive Text Classification by Combining GCN and BERT

Abstract

In this work, we propose BertGCN, a model that combines large scale pretraining and transductive learning for text classification. BertGCN constructs a heterogeneous graph over the dataset and represents documents as nodes using BERT representations. By jointly training the BERT and GCN modules within BertGCN, the proposed model is able to leverage the advantages of both worlds: large-scale pretraining which takes the advantage of the massive amount of raw data and transductive learning which jointly learns representations for both training data and unlabeled test data by propagating label influence through graph convolution. Experiments show that BertGCN achieves SOTA performances on a wide range of text classification datasets. Code is available at https://github.com/ZeroRin/BertGCN.

Accepted by Findings of ACL 2021

References in corpus (6)

• Semi-Supervised Classification with Graph Convolutional Networks
• Simplifying Graph Convolutional Networks
• Convolutional Neural Networks for Sentence Classification
• Variational Graph Auto-Encoders
• Uncertainty-aware Self-training for Text Classification with Few Labels
• Text Level Graph Neural Network for Text Classification

Cited by in corpus (4)

• Vul-LMGNNs: Fusing language models and online-distilled graph neural networks for code vulnerability detection
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• GNN-LM: Language Modeling based on Global Contexts via GNN
• $k$Folden: $k$-Fold Ensemble for Out-Of-Distribution Detection