摘要
针对目前成绩预测方法中存在准确率不高、实施性不强、可行性不佳等问题,文章提出一种基于浅层神经网络的预测模型。该模型采用调整的共轭梯度优化算法,将输入层与输出层进行连接,然后应用感知器进行学习成绩预测。与传统方法相比,该预测方法精度和准确率更高,而且实用性更强,能为后续优化与发展网络在线教育提供参考。
To solve such problems as low accuracy,low implementability and feasibility of the current performance prediction methods,this paper,based on Shallow Neural Network(SNN),proposes a prediction model.The model uses an adjusted conjugate gradient optimization algorithm to connect the input layer with the output layer,and then applies the perceptron for learning performance prediction.Compared with traditional method,the prediction method in this paper has higher precision and accuracy,and is more practical,which thus guarantees itself a reference for the subsequent optimization and development of online education.
作者
冯广
罗时强
陈卓
江家懿
伍文燕
Guang FENG;Shiqiang LUO;Zhuo CHEN;Jiayi JIANG;Wenyan WU(School of Automation,Guangdong University of Technology,Guangzhou Guangdong 510006;School of Computer Science,Guangdong University of Technology,Guangzhou Guangdong 510006;Network Information and Modern Education Technology Center,Guangdong University of Technology,Guangzhou Guangdong 510006)
出处
《中国教育信息化》
2022年第8期86-94,共9页
Chinese Journal of ICT in Education
基金
2017年国家自然科学基金“电子商务交互式决策助手对用户购物决策行为的影响与演化研究”(编号:71671048)
2020年中国高校产学研创新基金“新一代信息技术创新项目”(编号:2020ITA02013)。
关键词
浅层神经网络
优化算法
成绩预测
在线教育
Shallow neural network
Optimization algorithm
Grade prediction
Online education
