An Improved Neural Network Model for Graduate Education Evaluation

As an important part of higher education evaluation, graduate education evaluation plays a significant role to safeguard a sustainable development of higher education, especially graduate education. Considering the problems that traditional evaluation methods mainly depend on the experience of experts to a certain extent and there are many correlative factors which make it difficult to establish the evaluation model, a novel multiple improved PIDNN model is proposed in this paper to get the education evaluation model. In this model, the concepts of variable integral and partial differential are introduced into the design of hidden-layer of PIDNN, and the multiple improved PIDNN are dynamically combined to get the evaluation output with a gating network. The simulation results with the real data provided by East China University of Science and Technology indicate the validity of this modeling approach. Keywords -- education evaluation; PID neural network (PIDNN); variable integral; partial differential; multi-model