Abstract:

Because of the lack of disease awareness in depressed patients and the lack of early screening methods, most patients had developed to major depressive disorder when they were first diagnosed with depression. In order to improve the current situation, machine learning has been gradually used in some aspects of depression recently years, including early prediction, early recognition, auxiliary diagnosis, and treatment. In the application, the factors that affect the accuracy of machine learning model include the type and size of sample set, feature engineering, algorithm type, etc. In the future, machine learning should be further integrated into the health care system and mobile applications, continuously optimizing the machine learning model, fully mining patient health data to improve depression-related problems in terms of the prevention, identification, diagnosis, treatment and so on.

Key words: machine learning, depression, prediction, algorithm, model