The First Multilingual Model For The Detection of Suicide Texts

TL;DR

This paper introduces a multilingual transformer-based model to detect suicidal texts across six languages, demonstrating high accuracy and emphasizing the importance of linguistic diversity in mental health detection tools.

Contribution

It presents the first multilingual model using transformer architectures for suicide text detection across multiple languages, leveraging translation and fine-tuning techniques.

Findings

mT5 achieved F1 scores above 85% across languages

High-quality translations confirmed by perplexity measures

Multilingual models improve cross-lingual suicide risk detection

Abstract

Suicidal ideation is a serious health problem affecting millions of people worldwide. Social networks provide information about these mental health problems through users' emotional expressions. We propose a multilingual model leveraging transformer architectures like mBERT, XML-R, and mT5 to detect suicidal text across posts in six languages - Spanish, English, German, Catalan, Portuguese and Italian. A Spanish suicide ideation tweet dataset was translated into five other languages using SeamlessM4T. Each model was fine-tuned on this multilingual data and evaluated across classification metrics. Results showed mT5 achieving the best performance overall with F1 scores above 85%, highlighting capabilities for cross-lingual transfer learning. The English and Spanish translations also displayed high quality based on perplexity. Our exploration underscores the importance of considering linguistic diversity in developing automated multilingual tools to identify suicidal risk. Limitations exist around semantic fidelity in translations and ethical implications which provide guidance for future human-in-the-loop evaluations.