基于AHP-BP神经网络的高校实践教学评价研究

Research on Practical Teaching Evaluation of Colleges and Universities Based on AHP-BP Neural Network

以吉林建筑科技学院实践课程为样本,在“新工科”背景下确定13项评价指标,建立高校实践教学评价体系。利用层次分析法构建层次模型,确定分项指标权重。以三级指标为输入样本,一级指标为输出样本构建BP神经网络。运用MATLAB实现神经网络训练,SIM函数进行仿真预测。通过对期望输出与预测输出的误差计算,确定该模型能够在可接受范围内迅速输出评价结果,简化了单一运用AHP评价的计算过程。对高校实践类课程的评价工作提供一定的研究基础。

Taking the practice course of Jilin Institute of Architecture and Technology as a sample,13 evaluation indexes are determined under the background of"new engineering",and the evaluation system of college practice teaching is established.AHP is used to build a hierarchical model and determine the weight of sub-index.The BP neural network was constructed with the three-level index as input sample and the first-level index as output sample.MATLAB neural network training,SIM function simulation prediction.By calculating the error between the expected output and the predicted output,it is confirmed that the model can output the evaluation results rapidly in the acceptable range,and simplifies the calculation process of single AHP evaluation.It provides a certain research basis for the evaluation of practical courses in colleges and universities.