A hybrid model for student grade prediction using support vector machine and neural network

It is difficult for the intelligent teaching system in colleges to effectively predict student grade, which makes it difficult to formulate follow-up teaching strategies. In order to improve the effect of student grade prediction, this study improves the neural network algorithm, combines support vector machines to build a student grade prediction model, and uses PCA to reduce the dimensionality of the sample data. The specific operation is realized by SPSS software. Moreover, this study removes redundant information inside the input vector and compresses multiple features into a few typical features as much as possible. In addition, the research set a control experiment to analyze the performance of the research model and compare the advantages and disadvantages of the classification prediction effect of traditional machine learning algorithms and neural network algorithms. Through experimental comparison, we can see that the model constructed in this paper has certain advantages in all aspects of parameter performance, and the prediction model proposed in this study has certain effects.