Sentence-Level Grammatical Error Identification as Sequence-to-Sequence Correction

TL;DR

This paper presents an attention-based sequence-to-sequence model for sentence-level grammatical error identification and correction, demonstrating superior performance on the AESW 2016 Shared Task using a combination of character and word-based models.

Contribution

It introduces a novel combination of character-based and word-based encoder-decoder models with CNNs for improved grammatical error detection and correction.

Findings

Character-based models outperform word-based models.

The combined model achieves the highest accuracy on AESW 2016.

Sequence-to-sequence models can effectively identify and correct grammatical errors.

Abstract

We demonstrate that an attention-based encoder-decoder model can be used for sentence-level grammatical error identification for the Automated Evaluation of Scientific Writing (AESW) Shared Task 2016. The attention-based encoder-decoder models can be used for the generation of corrections, in addition to error identification, which is of interest for certain end-user applications. We show that a character-based encoder-decoder model is particularly effective, outperforming other results on the AESW Shared Task on its own, and showing gains over a word-based counterpart. Our final model--a combination of three character-based encoder-decoder models, one word-based encoder-decoder model, and a sentence-level CNN--is the highest performing system on the AESW 2016 binary prediction Shared Task.