结合主成分分析的BP神经网络模型的教学质量评价方法研究

Teaching Quality Evaluation Method Based on Principal Component Analysis and BP Neural Network Model

提出结合主成分分析PCA(principal component analysis)的BP神经网络模型的教学质量评价方法,首先利用主成分分析PCA对多维度教学质量评价指标进行了降维,确保了多维评价指标作为整体进行BP模型训练和测试,然后构建BP神经网络教学质量评价模型,利用其前向反馈计算和误差后向传播的特性,自适应获取评价指标之间的权值。实验证明,该方法可行,解决了传统评价方法的复杂建模问题,避免人为的主观随意度,保证评估结果的有效性。

This paper presents a teaching quality evaluation method based on principal component analysis and BP neural network model.First,PCA is used to reduce the dimension of teaching quality evaluation indexes,and ensure the multi-dimensional evaluation index used as a whole to train and test the BP model.Then a BP neural network model for teaching quality evaluation is constructed,and the weight of indexes and threshold of evaluation results are adaptively obtained by using the characteristics of forward feedback calculation and error backward propagation.The experiments show that the method is feasible.It solves the complex modeling problem in the traditional evaluation method and ensures the effectiveness and objectivity of the evaluation results.