Classifying Syntactic Errors in Learner Language

TL;DR

This paper introduces a cross-linguistic method for classifying syntactic errors in learner language using Universal Dependencies, aiding error analysis and evaluation of grammatical correction systems.

Contribution

The novel approach applies Universal Dependencies for syntactic error classification across languages, enhancing analysis of learner errors and GEC system outputs.

Findings

Effective classification of syntactic errors in English and Russian

Provides detailed error profiles for learner language

Useful for evaluating GEC system performance

Abstract

We present a method for classifying syntactic errors in learner language, namely errors whose correction alters the morphosyntactic structure of a sentence. The methodology builds on the established Universal Dependencies syntactic representation scheme, and provides complementary information to other error-classification systems. Unlike existing error classification methods, our method is applicable across languages, which we showcase by producing a detailed picture of syntactic errors in learner English and learner Russian. We further demonstrate the utility of the methodology for analyzing the outputs of leading Grammatical Error Correction (GEC) systems.