一种改进的BP神经网络在教学过程中的应用

Application of teaching process about improved BP neural network

利用一种改进的BP神经网络(PSO-BP)建立的教学成绩给定模型,在教学过程中所涉及的平时成绩给定的指标体系基础上,利用给定指标值作为输入,成绩估算值作为输出,通过PSO对BP神经网络的参数进行训练和学习,并利用Matlab软件建立实验平台。实验结果表明:新算法充分利用了神经计算的快速性以及粒子群算法的全局寻优能力,使得模型具有良好的辨识精度,可以较好地解决教学过程中平时成绩给定的动态问题。

A teaching achievement given model was established based on improved BP neural network （PSO-BP）.In the basis of achievement given indicator system during teaching process,the index value are input and achievement estimates are output,PSO was applied to the parameter learning and training of BP neural network.Experimental results of Matlab simulation showed that the new arithmetic did a better job which make the best of faster neural computation and PSO’s global optimization ability,and making the model has good identification accuracy,it can effectively solve the dynamic problem of regular achievement given in teaching process.