This repository contains a partial implementation of the code corresponding to the research paper titled "Harnessing the Power of Hugging Face Transformers for Predicting Mental Health Disorders in Social Networks".
Early diagnosis of mental disorders and intervention can facilitate the prevention of severe injuries and the improvement of treatment results. This study uses social media and pre-trained language models to explore how user-generated data can predict mental disorder symptoms. Our study compares four different BERT models of Hugging Face with standard machine learning techniques used in automatic depression diagnosis in recent literature. The results show that new models outperform the previous approach with an accuracy rate of up to 97%. Analyzing the results while complementing past findings, we find that even tiny amounts of data (Like users’ bio descriptions) have the potential to predict mental disorders. We conclude that social media data is an excellent source of mental health screening, and pre-trained models can effectively automate this critical task.
@ARTICLE{10438433,
author={Pourkeyvan, Alireza and Safa, Ramin and Sorourkhah, Ali},
journal={IEEE Access},
title={Harnessing the Power of Hugging Face Transformers for Predicting Mental Health Disorders in Social Networks},
year={2024},
volume={12},
number={},
pages={28025-28035},
keywords={Social networking (online);Mental health;Depression;Transformers;Task analysis;Data models;Predictive models;Machine learning;Text mining;Clinical diagnosis;Injuries;Machine learning;mental health;social networks;text mining;transformers},
doi={10.1109/ACCESS.2024.3366653}}
Please note that this repository only includes a partial implementation of the code described in the research paper. It is intended to provide a reference and starting point for interested researchers. To access the complete dataset used in the research, please follow the link here.