Unmasking digital despair: A language model framework for predicting depression on social media

The study aimed to develop a deep learning model to predict depression using Twitter/X data. Researchers analyzed tweets from April 2023 to July 2024 and identified 2,275 depressive users and 1,661 non-depressive users. The RoBERTa model achieved high performance with an accuracy of 0.822, an F1 score of 0.855, and an AUC of 0.809. Using the GPT-4o model, the main causes of depression were identified, including societal pressure, low self-esteem, cultural influences, and identity-related challenges. The findings demonstrate the potential of deep learning models for early depression screening using social media data. These insights may contribute to the development of targeted prevention strategies and improved mental health support.