Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Previously submitted to: JMIR Mental Health (no longer under consideration since Jul 13, 2025)

Date Submitted: Jul 11, 2025

Warning: This is an author submission that is not peer-reviewed or edited.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.

Precise Identification of User Needs and Service Optimization in Online Mental Health Service Platforms: A Text Mining-Based Empirical Analysis

  • Wei Wu; 
  • Longyuan Jiang; 
  • Yuanqi Xu; 
  • Lefei Fang; 
  • Shujing Dai; 
  • Jiakun Tian; 
  • Haiyan Yu

ABSTRACT

Background:

Online mental health service platforms, as a form of digital intervention, are increasingly pivotal in the field of mental health. This study aims to accurately identify user needs through text mining techniques to support needs-based decision-making and optimize mental health care services.

Objective:

The objective of this study is to accurately identify user needs on online mental health service platforms through advanced text mining techniques, including Latent Dirichlet Allocation (LDA) topic modeling, sentiment analysis, and semantic network analysis, and to quantitatively analyze these needs by integrating LDA with a need-intensity formula. By doing so, the study aims to address the challenges faced by current online mental health platforms—such as structural imbalances, insufficient cultural and individual sensitivity, and delayed responses to dynamic needs—and provide data-driven support and targeted optimization suggestions for improving the effectiveness and accessibility of these platforms, thereby better meeting modern society’s mental health service demands.

Methods:

We collected user comments from online mental health platforms using Python-based web scraping. After data cleaning, Chinese word segmentation, and removal of stop words, we employed a Latent Dirichlet Allocation (LDA) model to extract latent topics from the unstructured text. We also used ROST Content Mining System (Version 6.0) software for sentiment analysis and semantic network analysis of the topics. Furthermore, we integrated LDA with a need-intensity formula to quantitatively analyze user needs, enabling a deeper exploration of key needs and their emotional tendencies.

Results:

A total of 25,131 user comments were collected, and six main latent topics were identified: “Family and Marriage Counseling,”“Social and Emotional Support” and more. Sentiment analysis revealed 77.39% negative sentiment. The keywords “emotion”, “family”, “study” and “friends” in the semantic network highlighted the needs for emotion management, family relationship, academic pressure and social support needs. The analysis of the intensity of needs shows that “career and personal growth”(0.94) and “social and emotional support”(0.88) are the most urgent needs.

Conclusions:

Users express strong service needs on online mental health platforms, especially regarding emotion management and counseling for psychological disorders. Emotional support in the context of family relationships and social interaction is also a major focus of user concern. Online mental health service platforms should optimize their services by enhancing professional support and adding features for family relationship management and social interaction to better meet the diverse needs of users.


 Citation

Please cite as:

Wu W, Jiang L, Xu Y, Fang L, Dai S, Tian J, Yu H

Precise Identification of User Needs and Service Optimization in Online Mental Health Service Platforms: A Text Mining-Based Empirical Analysis

JMIR Preprints. 11/07/2025:80536

DOI: 10.2196/preprints.80536

URL: https://preprints.jmir.org/preprint/80536

PDF not available

The author of this paper has made a PDF available, but requires the user to login, or create an account.

© 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.