线上教学智能学业预警和精准干预研究

Research on Intelligent Education Warning and Precise Intervention in Online Teaching

智能学业预警是教育大数据、云计算等智能技术应用的热点,许多研究采用数据挖掘、机器学习、深度学习等技术来预测成绩.在线上教学常态化的背景下,智能学业预警和精准干预为下一步混合教学和职业教育的“三教”改革决策提供实证,也是教育大数据的重要研究内容.本研究设计了基于随机森林的预测模型分析教师和学生行为,并评估教师和学生行为中的“强特征”,最终实现针对学业预警中的“强特征”采用干预措施.使用制作的训练数据和验证数据进行实验,结果表明本研究模型训练数据准确率达到96%以上,验证数据准确率达到90%以上,学业预警精度高,同时针对学生行为和教师行为中的“强特征”制定了精准干预措施.

Intelligent study warning is a hot spot in the application of big data of education,cloud computing and other intelligent technologies.Many researches use data-mining,machine learning,deep learning and oth-er technologies to predict performance models.Under the background of online teaching normalization,intelli-gent study warning and precise intervention provide empirical evidence for the next educational reform deci-sionmaking of mixed teaching and vocational education,which is also an important research content of big data.In this study,we design a prediction model based on Random Forest to analyze the behavior of teachers and students,and evaluate the“strong features”of teachers and students,and finally achieve the interven-tion measures for the“strong features”.Using the training data and validation data,the results indicate that the research model is effective;the accuracy of training data is more than 96%;the accuracy of verification data is more than 90%,and the precision of model is high.At the same time,precise intervention measures are formulated for the“strong features”of students’and teachers’behaviors.