Design and Optimization of University Human Resource Management System Based on BP Neural Network Algorithm

In HRM(Human resource management), it has become an inevitable trend to integrate human resources, improve management methods and accelerate the overall development of human resources through information technology. In this paper, the design and optimization of university HRM system based on BPNN (BP neural network) algorithm are studied. Establish HRM system and apply three-layer BPNN model to salary forecast. In the actual budget, the standard BP algorithm has two important problems: slow convergence speed and easy to fall into local extremum. Therefore, this paper improves the BP algorithm by adding momentum term. The results show that the risk output value is in good agreement with the expert evaluation score, and the error is basically below 5%. The salary forecasting model optimized by NAG has a fast convergence rate and good forecasting effect, and the accuracy rate is not low, so it has certain practicability and reference value.