Computational Model-Based Prediction of Mental Health Problems and Optimization of Automated Intervention Strategies

Abstract

The traditional mental health assessment is highly subjective, and it is difficult to accurately assess the people who suffer from mental illnesses and treat the people who suffer from mental problems. This paper firstly analyzes the formation of mental health and the factors affecting people's mental health, and gets the way and standard of measuring mental health. Secondly, it collects and investigates people's daily behaviors and associates the extracted behavioral features with mental health. Finally, a prediction model was built based on the associated features, and people with mental illness were predicted based on BP neural network, and psychotherapeutic interventions were provided to them. The results show that at the 14th set of test data, the predicted and actual values are 111 and 111.5 respectively, which are less different. And the strategy optimization time of the model is 10ms, which indicates that the response time of the predictive model is shorter, which can process the predicted data quickly and reduce the memory of the model. The proposed study is able to detect mental health problems more quickly and accurately than other methods and solve mental health problems in time.