cantnlp@LT-EDI-2024: Automatic Detection of Anti-LGBTQ+ Hate Speech in Under-resourced Languages

TL;DR

This paper presents a transformer-based system for detecting anti-LGBTQ+ hate speech across ten under-resourced languages, incorporating script-switching data to improve multilingual social media comment classification.

Contribution

The study introduces a multilingual classification model with domain adaptation using synthetic and organic script-switched data for under-resourced languages.

Findings

Ranked second for Gujarati and Telugu

Script-switching data improves language detection

Performance varies across different languages

Abstract

This paper describes our homophobia/transphobia in social media comments detection system developed as part of the shared task at LT-EDI-2024. We took a transformer-based approach to develop our multiclass classification model for ten language conditions (English, Spanish, Gujarati, Hindi, Kannada, Malayalam, Marathi, Tamil, Tulu, and Telugu). We introduced synthetic and organic instances of script-switched language data during domain adaptation to mirror the linguistic realities of social media language as seen in the labelled training data. Our system ranked second for Gujarati and Telugu with varying levels of performance for other language conditions. The results suggest incorporating elements of paralinguistic behaviour such as script-switching may improve the performance of language detection systems especially in the cases of under-resourced languages conditions.