Research on the optimal allocation of resources for innovation and entrepreneurial education in universities in the context of deep learning

Abstract

Abstract Optimizing educational resource allocation will undoubtedly influence how universities construct their entrepreneurship and innovation curricula. In this study, we examine innovation and entrepreneurial education, allocate educational resources, and choose neural network methods related to deep learning. The convolutional neural network algorithm is studied in four stages: the convolutional layer, the pooling layer, the activation function, and the fully connected layer. The convolutional neural network model uses backpropagation to adjust the output parameters’ divergence from the ideal values, adjust and update the weight parameters, and confirm the computational layer data and hidden layer data from the propagation process. The DEA algorithm is reviewed to enhance and evaluate the resource allocation for innovation and entrepreneurial education at universities, and a DEA-BPNN efficiency assessment technique is created. The complete quality of students has a strong positive association with the effectiveness of allocating resources for entrepreneurship and innovation education. The efficiency of deploying resources for innovation and entrepreneurial education increases by 0.0512 per unit improvement in student quality.