Every Document Owns Its Structure: Inductive Text Classification via Graph Neural Networks

Abstract

Text classification is fundamental in natural language processing (NLP), and Graph Neural Networks (GNN) are recently applied in this task. However, the existing graph-based works can neither capture the contextual word relationships within each document nor fulfil the inductive learning of new words. In this work, to overcome such problems, we propose TextING for inductive text classification via GNN. We first build individual graphs for each document and then use GNN to learn the fine-grained word representations based on their local structures, which can also effectively produce embeddings for unseen words in the new document. Finally, the word nodes are aggregated as the document embedding. Extensive experiments on four benchmark datasets show that our method outperforms state-of-the-art text classification methods.

To appear at ACL 2020

References in corpus (2)

• Fi-GNN: Modeling Feature Interactions via Graph Neural Networks for CTR Prediction
• GraphAIR: Graph Representation Learning with Neighborhood Aggregation and Interaction

Cited by in corpus (7)

• A Survey on Text Classification: From Shallow to Deep Learning
• Graph Neural Networks for Natural Language Processing: A Survey
• Enhancing Keyphrase Extraction from Academic Articles Using Section Structure Information
• Dynamic Graph Collaborative Filtering
• Transformer-F: A Transformer network with effective methods for learning universal sentence representation
• Graph Convolutional Network with Generalized Factorized Bilinear Aggregation
• Revisiting SVD to generate powerful Node Embeddings for Recommendation Systems