Predicting MOOC Dropout over Weeks Using Machine Learning Methods 论文

摘要

With high dropout rates as observed in many current larger-scale online courses, mechanisms that are able to predict stu-dent dropout become increasingly impor-tant. While this problem is partially solved for students that are active in online fo-rums, this is not yet the case for the more general student population. In this pa-per, we present an approach that works on click-stream data. Among other features, the machine learning algorithm takes the weekly history of student data into ac-count and thus is able to notice changes in student behavior over time. In the later phases of a course (i.e., once such his-tory data is available), this approach is able to predict dropout significantly better than baseline methods.