典型文献
TELF-PSPOC:Three-layer Ensemble Learning Framework for Predicting Student Performance of Online Courses
文献摘要：
Virtual learning environment(VLE)MOOC provides large-scale data of resources,activities,and interactions within a course structure for predicting student performance.But it is challenging to extract and learn efficient features from student behaviors.In this paper,a three-layer ensemble learning framework for predicting student performance of online courses(TELF-PSPOC)at an early phase is proposed to analyze data collected from Open University Learning Analytics Dataset(OULAD).First,feature augmentation of student behavior is proposed to enrich current features of student performance,including pass rate and grades of all staged tests,daily clicks of online resources.Second,three-layer ensemble feature learning with heterogeneous classifiers(TEFL-HC)is proposed to benefit the integration of tree model and neural network.Compared with current two-layer ensemble learning,pretraining of features prevents overfitting while using nonlinear regression.The experiment shows that our TELF-PSPOC performs better than several baseline models.Besides,the relationship of the learning results and student behavior via VLE is further discovered.
文献关键词：
中图分类号：
作者姓名：
Kun Ma;Nan Zheng;Shan Jing;Zhenxiang Chen;Bo Yang
作者机构：
School of Information Science and Engineering,University of Jinan,Jinan,250022,China
文献出处：
引用格式：
[1]Kun Ma;Nan Zheng;Shan Jing;Zhenxiang Chen;Bo Yang-.TELF-PSPOC:Three-layer Ensemble Learning Framework for Predicting Student Performance of Online Courses)[J].计算机教育,2022(12):83-93
A类：
TELF,PSPOC,OULAD,TEFL
B类：
Three,layer,Ensemble,Learning,Framework,Predicting,Student,Performance,Online,Courses,Virtual,learning,environment,VLE,MOOC,provides,large,scale,data,resources,activities,interactions,within,structure,predicting,student,performance,But,challenging,extract,efficient,features,from,behaviors,In,this,paper,three,ensemble,framework,courses,early,phase,proposed,analyze,collected,Open,University,Analytics,Dataset,First,augmentation,enrich,current,including,pass,rate,grades,staged,tests,daily,clicks,Second,heterogeneous,classifiers,HC,benefit,integration,tree,neural,network,Compared,pretraining,prevents,overfitting,while,using,nonlinear,regression,experiment,shows,that,performs,better,than,several,baseline,models,Besides,relationship,results,via,further,discovered
AB值：
0.568595
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
