Teaching Quality Assessment Based on Principal Component Analysis and Elman Neural Network

Teaching quality assessment in higher education organizations is a complex nonlinear process in which various factors and variables are involved. Traditional assessment methods fail to reflect teaching quality with fairness and objectivity. This paper proposes a teaching quality assessment model based on principal component analysis (PCA) and Elman neural network. PCA was first used to reduce the dimensions of 12 original indices of an assessment system. 3 principal components were extracted as inputs of the Elman network to establish a PCA-Elman assessment model. The assessment performance of the proposed model was compared with a single Elman network model. The experiment results show that the structure of the PCA-Elman assessment model is simple, the convergence rate is fast, the assessment accuracy is high and the generalization ability is good.