摘要
为了对有无法通过科目考试风险的学生进行及时的干预、提升教学质量,开发学习预警系统。采用Matlab设计GUI界面实现平时成绩管理,为学习预警提供数据支持;采用三层BP神经网络实现期末成绩预测,输入特征量为23维归一化的平时成绩,隐含层和输出层的传递函数分别为tan-sigmoid和log-sigmoid。实验证明,该方法可有效筛选出需要预警的学生。
In order to intervene students at risk of failing and improve the teaching quality,the learning early warning system is developed.Matlab GUI is adapted to design the system interface and provide data support for learning early warning.To achieve final performance prediction,3-layer BP neural network is adapted,in which normalized 23-dimensional normal score data is used as input features,and tan-sigmoid and log-sigmoid transfer functions is designed for the hidden layer and output layer respectively.It is proved that this method can effectively screen out the students who need early warning.
作者
朱春媚
郑锦南
Zhu Chunmei;Zheng Jinnan(Mechanical and Electrical Engineering College,Zhongshan Institute,University of Electronic Science and Technology of China,Zhongshan 528400,China)
出处
《现代计算机》
2023年第19期117-120,共4页
Modern Computer
基金
广东省高等教育科学研究项目(2021GXJK317)。
关键词
MATLAB
神经网络
成绩预测
学习预警
Matlab
neural network
performance prediction
learning early warning
