Understanding Graph Convolutional Networks for Text Classification

TL;DR

This paper provides a comprehensive analysis of how graph construction techniques and GCN learning mechanisms impact text classification performance, offering new insights into their roles and effectiveness.

Contribution

It is the first study to systematically analyze node and edge embedding choices and GCN training strategies in various text classification benchmarks.

Findings

Node and edge construction significantly affect GCN performance.

Different GCN training mechanisms have varying impacts on classification accuracy.

Insights guide better graph construction for NLP tasks.

Abstract

Graph Convolutional Networks (GCN) have been effective at tasks that have rich relational structure and can preserve global structure information of a dataset in graph embeddings. Recently, many researchers focused on examining whether GCNs could handle different Natural Language Processing tasks, especially text classification. While applying GCNs to text classification is well-studied, its graph construction techniques, such as node/edge selection and their feature representation, and the optimal GCN learning mechanism in text classification is rather neglected. In this paper, we conduct a comprehensive analysis of the role of node and edge embeddings in a graph and its GCN learning techniques in text classification. Our analysis is the first of its kind and provides useful insights into the importance of each graph node/edge construction mechanism when applied at the GCN training/testing in different text classification benchmarks, as well as under its semi-supervised environment.