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{183752,
        author = {ANKIT ANUPAM ROUT and Deepanjal Sood and Ishmeet Singh},
        title = {Multimodal AI Framework for Early Detection of Mental Health Disorders via Social Media Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {2833-2840},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183752},
        abstract = {Mental health problems are growing in number, which has highlighted the necessity to have scalable and real-time detection systems within mental health. Social media are a fertile ground of emotionality and, therefore, beneficial avenues of passive mental health surveillance. The proposed deep learning framework of multi-label emotion classification over social media textual data is represented in this paper based on social media text data. The model was able to find the combination of several co-occurring emotions in a single post with good success using a Bidirectional Gated Recurrent Unit (BiGRU). The suggested system has tokenized text sequences, embedded vectors, and a sigmoid-activated output layer to predict the presence of eleven emotion reputations. The model is tested on preciseness, recall, F1-score, ROC and precision-recall curves, which shows high performance in several categories of emotion. This is a potentially fruitful early mental health surveillance and real-time emotional analysis.},
        keywords = {},
        month = {August},
        }