Copyright © 2025 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{184864,
        author = {D SRAVANTHI and D.MURALI},
        title = {PREDICTION AND DETECTION OF FUTURE MENTAL DISORDERS USING SOCIAL MEDIA DATA},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {4},
        pages = {3216-3221},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=184864},
        abstract = {Mental disorders affect millions of people worldwide, significantly impacting their thoughts, emotions, and behaviours. These conditions often go undetected until they become severe, making early detection a critical challenge. Timely identification, however, opens the door to early intervention, which can help individuals receive support before their condition worsens. One promising approach to address this challenge involves analysing how people express themselves on social media. These platforms have become digital diaries, where users often share their feelings, experiences, and emotional states. By examining these emotional cues, researchers can uncover patterns that may indicate underlying mental health conditions. In this study, we focus on two computational representations aimed at capturing emotional presence and variability in social media posts. These representations help in modelling not just what users feel, but also how their emotions change over time. We tested our models using two recent public datasets focused on Depression and Anorexia. The findings reveal that emotional signals extracted from user posts can effectively highlight individuals at risk. When both representations are combined, the model’s performance improves significantly. It matches the top-performing method for depression detection and trails the best anorexia model by just 1%.A key benefit of these emotional models is their interpretability. Unlike black-box deep learning methods, they offer clearer, more understandable insights.},
        keywords = {Mental health, Social media analysis, Depression detection, Anorexia detection, Emotion variability, Computational representation, Early detection, Interpretability, Natural Language Processing (NLP), Machine Learning},
        month = {September},
        }