Abstract:

Mental health disorders such as depression, anxiety, stress, and bipolar disorder are rapidly increasing due to lifestyle changes and digital dependency. Social media platforms have become an important medium where users openly express emotions, thoughts, and behavioral patterns. This project presents a social media–driven mental disorder detection system that utilizes machine learning, ensemble learning, and large language models (LLMs) to analyze user-generated content and interaction behavior. The system extracts linguistic, emotional, and behavioral features from social media data and applies intelligent models to predict potential mental health conditions. By combining traditional machine learning with language understanding capabilities of LLMs, the proposed approach improves prediction accuracy and supports early identification, enabling timely intervention and mental health awareness.

Description:

.