关键词: 主成分分析, RBF神经网络, 成绩预测, BP神经网络

Abstract: To improve the accuracy of students’ writing score prediction, a prediction model based on principal component analysis (PCA) and radial basis function (RBF) was proposed. First the dimensions of an established assessment system of students’ writings were reduced. The first five principal components were extracted and taken as inputs of the RBF neural network. Then a three-layered RBF network prediction model was created. The experiment results show that compared with a simple RBF neural network and a BP neural network, the PCA-RBF prediction model is of simpler structure, faster convergence speed, higher prediction accuracy and better generalization ability. The effectiveness of the proposed model is verified.


Key words: principal component analysis, RBF neural network, score prediction, BP neural network