Attendance Data Based on BP Neural Network Model

Neural networks have better classification and memory capabilities, so the corresponding learning algorithms have become the focus of research. The purpose of this paper is to analysis of attendance data based on BP neural network model. The basic theory and technology are introduced, and the attendance rate of M University students is predicted by model, and compared with the actual number. In the process of solving and drawing conclusions, there is such a relationship between the predefined parameters of the neural network model and the required quantities, after selecting the initial value, the whole algorithm will solve the optimal value in a specific direction. In order to avoid errors caused by the selection of initial values and the creation of local minima, the verification data is used after microprocessing and continuous debugging of the program. The overall error of 10 students' attendance prediction fluctuates between 3 and 11. The parameters of training volume, training number, and learning rate in this paper are suitable for the prediction model of attendance.