Construction of real-time mental health early warning system based on machine learning

Abstract

Abstract With the great development of computer technology, it is feasible to apply robot learning to the prediction and diagnosis of mental illness. Establishing an online psychological health warning system helps psychologists to know the mental health status of patients and makes rapid intervention. Based on the principle of machine learning, the design idea of constructing online psychological early warning system were proposed here by using the patient’s daily performance and textual representation. The system included: client, server, doctor, and family. The client collected user text data and offered basic client technology; the server provided business logic of the web backend, database and mental health rating warning for client, doctor and family; the doctor mainly displayed the binding user status. When the mental health level of users was poor, early warning should be warning so that doctors can intervene in time. The relevant auxiliary treatment information for the patient would be send to family side, which help for the family therapy. This system provides a new way for doctors to monitor patients with mental illness, which but further study and training is needed to establish an effective model.