Prediction and Management of the Quality of Classified Student Training Based on an Improved Neural Network

Abstract

Personalized teaching and classified training promote each student to develop in the most suitable direction. The studies on classified student training (CST) mainly target the objects of CST, which refers to the classified training of underachievers and achievers, or the classified training of different groups of high-quality talents like doctoral students and master candidates. However, there are few research results on the implementation or quality forecast of CST. To fill up the gap, this paper explores the prediction and management of the CST quality based on an improved neural network. Firstly, a cognitive diagnosis model for CST was established to realize targeted group learning. Thereafter, an evaluation index system (EIS) was constructed for student learning quality. Next, a prediction model was built based on improved backpropagation neural network (BPNN), and the particle swarm optimization (PSO) was called to optimize the weights and thresholds of the neural network. The effectiveness of our model was proved through experiments. The relevant findings provide impetus to the timely update of CST.