| Abstract |
A significant rise in psychological issues, including symptoms of depression, anxiety, and disturbed sleep, has resulted from the worldwide quarantine measures imposed in reaction to the coronavirus epidemic. The world’s healthcare systems have been greatly burdened by these concerns. The most effective way to avoid depression is to identify the early warning signs of it. The healthcare business has recently shifted its emphasis to technology in order to pave the way for the future creation of sophisticated smart cities. There is hope that analyzing content on social media can significantly reduce depression and, by extension, suicide rates. Using a computational technique, this research demonstrates how to find depressed individuals on social media. The complex system finds text contents related to depression posts on online platforms using machine learning (ML) techniques. The research preprocessed and pre-trained three separate ML models using data from social media platforms. Differentiating between material that does not indicate depression and information that does is the main goal of this research. An empirical investigation found that the BERT model had the lowest validation loss and showed an accuracy rate of over 90%, suggesting optimum performance. Additionally, the evaluation results of all the performance metrics validate that the BERT model is robust and effective as compared with the two other ML models used in this study. This implies that the proposed scheme is capable of accurately classifying depressive-related posts across different social media platforms. © 2025 selection and editorial matter, Diptendu Sinha Roy, Mir Wajahat Hussain, K. Hemant Kumar Reddy, Deepak Gupta; individual chapters, the contributors. |
