Mental Health Risk Detection From Social Media Text Data: A Scoping Review of the Machine Learning Research Landscape
PsyCh Journal · 2026-06 · 15 (3)
DOI: 10.1002/pchj.70100 ↗Kurz gefasst
Scoping Review in der Zeitschrift PsyCh Journal, online erschienen am 15.05.2026. Verfasst von Yiqing He und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Psychology, Guangzhou University, Guangzhou, China. Hauptthemen laut MEDLINE: Social Media, Machine Learning, Mental Health, Mental Disorders. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Provincial Philosophy and Social Science Planning Project, Guangdong Provincial Department of Education Innovation-Driven University Strengthening Program for Youth Innovation Talent.
Abstract (Originalsprache)
Machine learning approaches have been increasingly applied to social media text data for mental health risk detection. However, existing studies vary widely in target outcomes, data sources, labeling strategies, and evaluation practices, and a structured overview of recent research remains limited. This scoping review aims to map the recent research landscape of machine learning–based mental health risk detection using social media text data. Following PRISMA ScR guidelines, peer reviewed journal articles published between January 2021 and January 2026 were retrieved from PubMed, Web of Science, and IEEE Xplore. Studies applying machine learning or deep learning methods to social media text data for mental health risk detection were included and synthesized descriptively. A total of 136 studies were identified. Most focused on depression, anxiety, and suicide or self‐harm related risks. Mental health risk was predominantly operationalized through proxy indicators derived from user‐generated content, with limited use of survey‐linked or clinically anchored labels. Traditional machine learning, deep learning, and Transformer‐based models coexisted, alongside substantial heterogeneity in validation strategies and performance metrics. Current research primarily targets proxy‐based mental health risk signals rather than clinical diagnoses. This review clarifies prevailing research emphases and methodological practices, and supports the use of social media–based approaches for population‐level monitoring and early risk identification.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Psychology, Guangzhou University, Guangzhou, China.
- Department of Artificial Intelligence, Faculty of Computer Science and Information Technology, University of Malaya, Kuala Lumpur, Malaysia.
ORCID-Profile: Yiqing He, Darong Liu
Förderung
- Guangdong Provincial Philosophy and Social Science Planning Project
- Guangdong Provincial Department of Education Innovation-Driven University Strengthening Program for Youth Innovation Talent
Zitieren
APA
He, Y., He, Y., & Liu, D. (2026). Mental Health Risk Detection From Social Media Text Data: A Scoping Review of the Machine Learning Research Landscape. PsyCh Journal, 15(3). https://doi.org/10.1002/pchj.70100
Vancouver
He Y, He Y, Liu D. Mental Health Risk Detection From Social Media Text Data: A Scoping Review of the Machine Learning Research Landscape. PsyCh Journal. 2026;15(3). doi:10.1002/pchj.70100
BibTeX
@article{He2026,
author = {He, Yiqing and He, Yinning and Liu, Darong},
title = {Mental Health Risk Detection From Social Media Text Data: A Scoping Review of the Machine Learning Research Landscape},
journal = {PsyCh Journal},
year = {2026},
volume = {15},
number = {3},
doi = {10.1002/pchj.70100}
}
EUVIMED