【摘要】 基于公有云以及微信小程序平台开发了一套云计算的高校学生学业预警系统。首先设计了云计算模式下的学业预警系统架构,阐述了网络端-管-云的信息交互关系。在预测算法方面,综合采用了LightGBM梯度提升算法和统计学诊断算法优化设计了LightGBM的参数。然后,通过数据增强方法并利用真实数据集开展毕业状态分类的实验分析和预警系统测试。结果表明,该技术的分类精度和计算效果高,且高等数学与毕业设计课程对毕业状态的影响度最高。目前,学生通过手机端就能提交成绩信息,也能够立即获得预警结果。更多还原

【Abstract】 The academic management of college students is an important content of college education management.Based on the public cloud and WeChat mini program platform,an early warning system is developed for college students’ academic work based on cloud computing.First,the framework for the academic early warning system is designed under the cloud computing mode,and the information interaction relationship among the network-management-cloud is explained.In terms of prediction algorithms,the LightGBM gradient boosting algorithm and statistical diagnosis algorithm are combined together,and the parameters of LightGBM are also optimized and well designed.Then,through the data augmentation method,the real data set is used to carry out the experimental analysis of graduation status classification and the early warning system test.The results show that it performs well in classification accuracy and calculation effect.Besides,two courses of the advanced mathematics and graduation design have the significant impact on graduation status.Nowadays,students who submit grade information through the mobile phone can also get the early warning results immediately.更多还原