SpellGCN: Incorporating Phonological and Visual Similarities into Language Models for Chinese Spelling Check

TL;DR

This paper introduces SpellGCN, a graph convolutional network that integrates phonological and visual similarities among Chinese characters into language models, significantly improving Chinese Spelling Check accuracy.

Contribution

It proposes a novel end-to-end trainable model that incorporates character similarity knowledge via a specialized graph convolutional network for Chinese Spelling Check.

Findings

Achieves superior performance on three datasets

Outperforms previous models by a large margin

Demonstrates effectiveness of phonological and visual similarity integration

Abstract

Chinese Spelling Check (CSC) is a task to detect and correct spelling errors in Chinese natural language. Existing methods have made attempts to incorporate the similarity knowledge between Chinese characters. However, they take the similarity knowledge as either an external input resource or just heuristic rules. This paper proposes to incorporate phonological and visual similarity knowledge into language models for CSC via a specialized graph convolutional network (SpellGCN). The model builds a graph over the characters, and SpellGCN is learned to map this graph into a set of inter-dependent character classifiers. These classifiers are applied to the representations extracted by another network, such as BERT, enabling the whole network to be end-to-end trainable. Experiments (The dataset and all code for this paper are available at https://github.com/ACL2020SpellGCN/SpellGCN) are conducted on three human-annotated datasets. Our method achieves superior performance against previous models by a large margin.