Diverse Perspectives, Divergent Models: Cross-Cultural Evaluation of Depression Detection on Twitter

Nuredin Ali Abdelkadir, Charles Zhang, Ned Mayo, Stevie Chancellor


Abstract
Social media data has been used for detecting users with mental disorders, such as depression. Despite the global significance of cross-cultural representation and its potential impact on model performance, publicly available datasets often lack crucial metadata relatedto this aspect. In this work, we evaluate the generalization of benchmark datasets to build AI models on cross-cultural Twitter data. We gather a custom geo-located Twitter dataset of depressed users from seven countries as a test dataset. Our results show that depressiondetection models do not generalize globally. The models perform worse on Global South users compared to Global North. Pre-trainedlanguage models achieve the best generalization compared to Logistic Regression, though still show significant gaps in performance on depressed and non-Western users. We quantify our findings and provide several actionable suggestions to mitigate this issue
Anthology ID:
2024.naacl-short.58
Original:
2024.naacl-short.58v1
Version 2:
2024.naacl-short.58v2
Volume:
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Kevin Duh, Helena Gomez, Steven Bethard
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
672–680
Language:
URL:
https://aclanthology.org/2024.naacl-short.58
DOI:
10.18653/v1/2024.naacl-short.58
Bibkey:
Cite (ACL):
Nuredin Ali Abdelkadir, Charles Zhang, Ned Mayo, and Stevie Chancellor. 2024. Diverse Perspectives, Divergent Models: Cross-Cultural Evaluation of Depression Detection on Twitter. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers), pages 672–680, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
Diverse Perspectives, Divergent Models: Cross-Cultural Evaluation of Depression Detection on Twitter (Abdelkadir et al., NAACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.naacl-short.58.pdf