Artificial Error Generation with Machine Translation and Syntactic Patterns

TL;DR

This paper introduces two novel methods for artificially generating writing errors—using machine translation and syntactic pattern extraction—to augment training data and improve automated error detection.

Contribution

It proposes treating error generation as a machine translation task and developing a pattern-based error insertion system, both enhancing error detection models.

Findings

Artificial error generation improves detection accuracy.

Machine translation approach effectively creates realistic errors.

Pattern-based method complements translation for diverse errors.

Abstract

Shortage of available training data is holding back progress in the area of automated error detection. This paper investigates two alternative methods for artificially generating writing errors, in order to create additional resources. We propose treating error generation as a machine translation task, where grammatically correct text is translated to contain errors. In addition, we explore a system for extracting textual patterns from an annotated corpus, which can then be used to insert errors into grammatically correct sentences. Our experiments show that the inclusion of artificially generated errors significantly improves error detection accuracy on both FCE and CoNLL 2014 datasets.