On the application of BP neural network based on Levenberg-Marquardt algorithm in the diagnosis of mental disorders

Because of the traditional BP neural network´s limitation, including easily falling into local minimum, slow convergence and iteration for a long time, we propose a diagnostic system of mental disorders based on Levenberg-Marquardt algorithm. This article illustrates the basic principles of the neural network algorithm and establishes a mental diagnosis model based on L-M algorithm which uses common mental illness among college students as subjects. The outcome of the simulation illustrates the fact that it is effective to apply the neural network to establish a mental diagnostic system because it has reduced the training iterations and training period and it has high accuracy as well.