Depression is one of the main reasons people commit suicide around the world. However, a significant portion of depressive cases remain misdiagnosed and untreated. Text chats are a useful tool for predicting stress because people spend the majority of their time online these days. This research proposes a MobiBART: Sentiment-Aware NLP Model Chatbot designed to predict and manage stress through textual interactions in real-time conversations. The model employs an End-to-End Workflow combining Sequence-to-Sequence (Seq2Seq) for stress feature extraction and a prediction Model to assess stress, providing effective intervention strategies. The system operates in two key phases. In Phase 1, users authenticate by registering on the platform with their necessary personal details users can log in to the system using their credentials. In Phase 2, the user can interact through introduced MobiBART chatbot where the model processes the conversation to detect stress level. It triggers notifications to the user’s pre-registered contacts if the model identifies signs of stress and also responds with comforting, empathetic messages to reduce the user’s stress. The dual functionality of system not only predicts stress but also offers proactive management solutions, empowering both users and their social circles to respond promptly to mental health concerns. This approach offers a comprehensive solution for stress detection and management through natural conversations by using advanced NLP techniques.
Balraj et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: