Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{176839,
        author = {Aananya Pawar and Maanasvi Mahajan and Priyanshi Joshi and Maria Jamal},
        title = {AI Enabled Mental Health Support Using Social Media Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {11},
        pages = {6683-6688},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=176839},
        abstract = {The increasing incidence of mental illness and widespread social media usage provide the means to identify psychological distress through digital clues. This paper envisions a machine learning system that searches user-created social media posts for evidence of probable symptoms of mental illness and provides referrals to proper sources. An openly available dataset of labelled mental health statements was pre-processed with regular NLP procedures, followed by feature extraction using BERT sentence embeddings for semantic richness. SMOTE was utilized to balance class representation against class imbalance. XGBoost was subsequently used to train a robust classifier, achieving overall accuracy of 87.73% for seven mental health classes. Other visualizations including t-SNE, confusion matrix, and heatmaps for classification also endorsed the robustness and interpretability of the model. Beyond its realm in classification, the system further includes an aid level by the personalized scheme in suggesting India-focused helplines on mental health derived from predicted emotions. The envisioned framework exemplifies that deep learning can be augmented with natural language processing analysis to promote scalable real-time mental treatment of health over online platforms.},
        keywords = {Mental Health, Natural Language Processing, BERT, XGBoost, SMOTE, Social Media, Emotion Classification},
        month = {April},
        }