X as a Mirror for Mental Health: The Role of AI in Analyzing Tweets for Depression

Depression is one of the most common and serious mental illnesses, posing a significant public health issue. Recent advances in artificial intelligence and machine learning opened the door to early detection through analyzing X tweets data. The paper discusses models such as LSTM and BERT, with a previous highest accuracy of 96.6%, and reports our enhanced results with AraBERT and MarBERT showing accuracies of 97.0% and 97.7% respectively. It also calls for reinforcing these systems and creating collaboration among mental health experts and AI researchers. The study aims to utilize artificial intelligence to shatter the stigma of seeking mental health assistance.