Research on the Optimization Strategy of Government Education Resource Allocation Based on Deep Learning

This paper focuses on optimizing the distribution of educational resources in the government and introduces the deep learning technique. This paper proposes a kind of Resource Assignment Neural Network (DWARN), which can accurately forecast the demand of resources and realize rational allocation. Through several channels, data from the educational department, field surveys, and online educational platforms are collected, including the location of the school, the population of students, the history of education funds, and so on. After cleaning and normalization preprocessing, the neural network is set to 0.001, 500, and 64 for experiments on the deep learning framework TensorFlow. Compared with the conventional linear programming, the DWARN algorithm achieves 90% precision in resource allocation, which is 25% better than the conventional algorithm. Experimental results show that DWARN can deal with complicated situations like educational policy changes. Finally, the paper puts forward the optimized resource allocation strategy, which can provide powerful support for the rational allocation of educational resources.