Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Mar 23, 2025
Open Peer Review Period: Mar 24, 2025 - May 19, 2025
Date Accepted: Nov 24, 2025
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Determinants of Inquiry Responses in Nonprofit Mental Health Forums: An Interpretable Machine Learning Approach
ABSTRACT
Background:
In recent years, there has been a noticeable increase in individuals seeking psychological assistance and therapeutic interventions through virtual mental health platforms. The increasing demands and contributions of users constitute pivotal elements underpinning the development of these online psychological communities. Nevertheless, limited attention has been given to the quality and quantity of information exchange within these virtual mental health forums.
Objective:
The primary objective of this study is to investigate the determinants influencing user inquiries and response volumes to queries within virtual mental health forums. This study aims to provide empirical data and theoretical underpinnings, as well as suggestions, to improve the service of online mental health communities.
Methods:
This study employs a pretrained deep learning-based natural language processing model named BERT to conduct a thematic analysis of the content in the inquiry and response sections of online mental health communities. In addition, sentiment analysis is performed based on a sentiment dictionary. We then utilize a machine learning method (LightGBM) for predictive analysis of the response volumes. Furthermore, the SHAP analytical technique is employed to better explain the determinants of response behavior.
Results:
The findings indicate that user inquiries can be categorized into seven themes: job, love, depressive moods, relationships, school, marriage, and family. The sentiment analysis reveals that topics related to 'relationships' and 'marriage' are associated with positive emotions and high view counts and contribute positively to the number of replies. Conversely, topics such as 'romance' and 'depression' usually carry neutral or adverse emotional tones and negatively impact response rates. Notably, posts with sufficiently high levels of positive or negative emotions tend to receive a higher number of responses.
Conclusions:
This study has both theoretical and practical implications for advancing the contribution of user knowledge in online mental health communities, as well as for improving the quality of services provided by these communities.
Citation
The author of this paper has made a PDF available, but requires the user to login, or create an account.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.