MOOC用户流失率预测模型研究
Research on the prediction of MOOC drop-out rate
随着互联网的发展，在线教育蓬勃发展。MOOC（massive open online courses）作为在线教育的一种形式，以其学习成本低等优点被越来越多的人接受。然而，过高的用户流失率却给MOOC的健康发展带来了极大的挑战。基于MOOC网站提供的39门课程的用户后台操作数据，首先对MOOC用户的行为模式进行分析，并对数据集进行特征提取和选择；然后，给出一种多分类器加权的用户流失率预测模型；接下来，提出可信集合概念，并结合多分类器加权模型给出一种带有可信集合的改进预测算法。最后，对所提的模型进行了验证，与单一分类模型相比较，多分类器加权模型进一步提高了模型的预测性能。
With the development of Internet, online education with various forms become more and more popular.. As a form of online education, MOOC has been accepted by more and more people for its low cost of learning. But the drop-out rate is too high for he sound development of MOOC . In this paper, a background dataset about 39 courses offered by MOOC website is used to model and analyze the behaviors of the learners . Then, a feature set is extracted from the dataset. On this basis, the paper proposed a multiple classifier weighted model to predict MOOC drop-out rate; Based on this, the concept of the classifier weight is put forward, and the paper gives an improved prediction algorithm with trusted set Finally, the proposed multiple classifier weighted model is verified. Compared with a single classification model, the proposed model make better prediction.
罗涛、刘文彦
教育计算技术、计算机技术
模式识别MOOC用户流失率分类与预测多分类器改进可信集合
pattern recognitionMOOCdrop-out rateclassification and predictionmultiple classifier weighted modeltrusted set
罗涛,刘文彦.MOOC用户流失率预测模型研究[EB/OL].(2016-01-12)[2025-08-23].http://www.paper.edu.cn/releasepaper/content/201601-247.点此复制
评论