Czech Grammar Error Correction with a Large and Diverse Corpus

TL;DR

This paper introduces a large, diverse Czech corpus for grammatical error correction, compares multiple GEC systems including Transformer-based models, and evaluates GEC metrics against human judgments to advance non-English language resources.

Contribution

It provides the first large Czech GEC corpus, benchmarks several GEC systems, and assesses metric reliability, filling a gap in non-English grammatical correction resources.

Findings

Transformer-based GEC systems perform competitively.

The new Czech corpus is publicly available for research.

Common GEC metrics show varying correlation with human judgments.

Abstract

We introduce a large and diverse Czech corpus annotated for grammatical error correction (GEC) with the aim to contribute to the still scarce data resources in this domain for languages other than English. The Grammar Error Correction Corpus for Czech (GECCC) offers a variety of four domains, covering error distributions ranging from high error density essays written by non-native speakers, to website texts, where errors are expected to be much less common. We compare several Czech GEC systems, including several Transformer-based ones, setting a strong baseline to future research. Finally, we meta-evaluate common GEC metrics against human judgements on our data. We make the new Czech GEC corpus publicly available under the CC BY-SA 4.0 license at http://hdl.handle.net/11234/1-4639 .