Grammatical Error Correction for Low-Resource Languages: The Case of Zarma
Domain:
natural language processingRecord type:
paperdataset
Creator:
KeiBreLe,Owu
Host:
Grammatical error correction (GEC) aims to improve text quality and readability. Previous work on the task focused primarily on high-resource languages, while low-resource languages lack robust tools. To address this shortcoming, we present a study on GEC for Zarma, a language spoken by over five million people in West Africa. We compare three approaches: rule-based methods, machine translation (MT) models, and large language models (LLMs). We evaluated GEC models using a dataset of more than 250,000 examples, including synthetic and human-annotated data. Our results showed that the MT-based approach using M2M100 outperforms others, with a detection rate of 95.82% and a suggestion accuracy of 78.90% in automatic evaluations (AE) and an average score of 3.0 out of 5.0 in manual evaluation (ME) from native speakers for grammar and logical corrections. The rule-based method was effective for spelling errors but failed on complex context-level errors. LLMs -- Gemma 2b and MT5-small -- showed moderate performance. Our work supports use of MT models to enhance GEC in low-resource settings, and we validated these results with Bambara, another West African language.
Similar
ChatGPT for Arabic Grammatical Error CorrectionEnhancing Text Editing for Grammatical Error Correction: Arabic as a Case StudyRobustness of Synthetic vs. Human-Annotated Grammatical Error Detection Models in Low-Resource LanguagesBeyond English: Evaluating LLMs for Arabic Grammatical Error CorrectionGrammatical Error Detection Performance in Low-Resource Languages: Zero-Shot Synthetic vs. Human-Annotated BaselinesDiversity in Zero-Shot Synthetic Data for Low-Resource Grammatical Error Detection