• Janginti Rishitha
• Boyala Niharika
• Kantha Ananya
• Dr B V Ramana Murthy

The rapid growth of social media has led to a huge increase in user generated textual data, which provides valuable insights into people’s mental health. Social media posts often reflect people’s emotions, stress, and inner concerns, making these platforms a useful source for the early detection of mental health disorders. However, many existing approaches emphasis on identifying a single disorder and fail to effectively capture both the contextual meaning and emotional intensity present in textual data. To address these limitations, this paper proposes a innovative deep learning framework called TMST-Net, a Multi-Task Sentiment-Gated Transformer network designed to detect multiple mental health disorders simultaneously. The proposed model utilizes a transformer-based encoder to extract contextual features from text and integrates a sentiment gating mechanism to highlight emotionally significant information. A multi-task learning strategy is employed to equally predict depression, anxiety, and burnout from social media posts. The model is trained on a dataset of over 30,000 social media posts and evaluated using standard classification metrics. The results indicate that TMST-Net effectively captures both contextual and emotional aspects of text, leading to improved performance in detecting mental health conditions. Additionally, a lightweight web application is developed to enable real-time mental health screening.

Mental Health Detection, Social Media Analysis, Transformer Model, Multi-Task Learning, Sentiment Analysis, Deep Learning, Natural Language Processing (NLP), Depression Detection, Anxiety Detection, Burnout Detection

"TMST-Net: A Multi-Task Sentiment-Gated Transformer Framework for Mental Health Detection from Social Media Text", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 3, page no.f723-f730, March-2026, Available :http://www.jetir.org/papers/JETIR2603591.pdf

"TMST-Net: A Multi-Task Sentiment-Gated Transformer Framework for Mental Health Detection from Social Media Text", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 3, page no. ppf723-f730, March-2026, Available at : http://www.jetir.org/papers/JETIR2603591.pdf