Year: 2025
Type: article
Abstract: The increasing prevalence of mental health issues worldwide has underscored the need for innovative, accessible diagnostic tools. Traditional mental health assessment methods often rely on self-reporting and professional evaluations, which is time-consuming and lack immediacy. Paper presents an AI-driven framework for the self-diagnosis of mental health status, leveraging machine learning models trained on a variety of user-generated data, including social media activity, profile characteristics, and social connections. By integrating feature engineering techniques, dimensionality reduction, and robust data preprocessing, the proposed model can detect and predict mental health states with considerable accuracy. The novelty of this work lies in its automated, real-time processing pipeline, which combines initial data validation and label assignment with advanced ML techniques for effective prediction. This framework offers users insights into their mental well-being and also serves as a scalable solution for mental health monitoring, potentially reducing the burden on healthcare systems by facilitating early intervention.
Source: Journal of Machine and Computing
Authors: Pallavi M O, Pundalik Chavan
Institution: REVA University
Topic: Mental Health via Writing
Subfield: Social Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4415312878