Nowadays people are facing various psychological problems. Existing solution of evaluation and treatment of mental illness is only to see a psychiatrist, but most of the users has sense of resistance on psychiatrist. Meanwhile most of the existing systems are psychological tests for entertainment, whose results cannot be accurately analyzed. On basis of this phenomenon, we develop the Baymax system to construct a relation between the potential patients' motion and a specific models in order to analyze the user's mental conditions and predict his or her potentialities for suffering certain mental diseases, by exerting and exploring the user's short messages, online diaries, and communicative information on social media, as well as the questionnaires accomplished by the user. Our system uses big data analysis method, and allows users to use the system to determine whether there is a possibility of mental illness in the absence of a psychiatrist.
