【摘要】 针对传统教学质量评价中存在的单一性和人为因素干涉等问题,为达到科学、准确的评价效果,本文提出了一种改进的小波神经网络教学质量评价方法,充分利用小波变换良好的局部化性质并结合神经网络的自学习、自适应能力,采用加动量批处理的小波神经网络算法进行网络训练,从大量评价数据中自动找出与评价结果之间的非线性关系,然后测试得到评价结果。实验证明,改进后的小波神经网络算法提高了收敛速度和准确性,使教学质量评价更客观、科学,具有一定的实用性和有效性。更多还原

【Abstract】 In view of the single factors and human interference and other issues in the traditional evaluation of teaching quality,in order to achieve scientific,accurate evaluation result,an improved wavelet neural network evaluation method of teaching quality is presented,it makes full use of the good localization property of wavelet transform,and combines with the self learning,adaptive capacity of neural network,adopting the momentum and batch processing wavelet neural network algorithm,training the network,to automatically identify with of the non-linear relationship between them and the evaluation results from a large number of evaluation data,and testing to get evaluation result.Experiments show that,the improved wavelet neural network algorithm improves the convergence speed and accuracy,making the teaching quality evaluation is more objective and scientific,which is practical and effective.更多还原