An Analysis of Language Frequency and Error Correction for Esperanto
Junhong Liang
TL;DR
This paper tackles the underexplored area of Grammar Error Correction for Esperanto by constructing two dedicated resources: the EO-GP corpus for frequency analysis and the EO-GEC dataset for authentic learner errors. It conducts a frequency-analytic study of Esperanto letters and words using La Eta Princo and the EO-GP data, validating Zipf-like word distributions and quantifying letter-entropy to compare with English. The core contribution is a comprehensive evaluation of GPT-3.5 and GPT-4 on Esperanto GEC, including automatic (ERRANT and M2Scorer) and human assessments, revealing that GPT-4 generally outperforms GPT-3.5, particularly in overall correction quality and POS-related errors. The results underscore the potential of large language models to address GEC for low-resource, constructed languages while highlighting data-volume limitations and the need for broader, richer Esperanto datasets to advance future research.
Abstract
Current Grammar Error Correction (GEC) initiatives tend to focus on major languages, with less attention given to low-resource languages like Esperanto. In this article, we begin to bridge this gap by first conducting a comprehensive frequency analysis using the Eo-GP dataset, created explicitly for this purpose. We then introduce the Eo-GEC dataset, derived from authentic user cases and annotated with fine-grained linguistic details for error identification. Leveraging GPT-3.5 and GPT-4, our experiments show that GPT-4 outperforms GPT-3.5 in both automated and human evaluations, highlighting its efficacy in addressing Esperanto's grammatical peculiarities and illustrating the potential of advanced language models to enhance GEC strategies for less commonly studied languages.