Say Anything: Automatic Semantic Infelicity Detection in L2 English Indefinite Pronouns

TL;DR

This paper investigates the challenges non-native English speakers face with indefinite pronouns, proposing and evaluating a deep learning-based method for automatically detecting subtle semantic errors in learner language.

Contribution

It introduces a linguistically motivated framework for semantic error detection in L2 English indefinite pronouns and demonstrates the effectiveness of deep learning models for this nuanced task.

Findings

Deep learning models show promise in detecting semantic anomalies.

Linguistic hypotheses about indefinite pronoun errors are empirically supported.

Automatic detection outperforms baseline methods.

Abstract

Computational research on error detection in second language speakers has mainly addressed clear grammatical anomalies typical to learners at the beginner-to-intermediate level. We focus instead on acquisition of subtle semantic nuances of English indefinite pronouns by non-native speakers at varying levels of proficiency. We first lay out theoretical, linguistically motivated hypotheses, and supporting empirical evidence on the nature of the challenges posed by indefinite pronouns to English learners. We then suggest and evaluate an automatic approach for detection of atypical usage patterns, demonstrating that deep learning architectures are promising for this task involving nuanced semantic anomalies.