Developing early warning systems to predict students’ online learning performance

作者：

Highlights：

• We develop early warning systems to predict at-risk students while a course is in progress.

• Learning portfolios from a fully online course are evaluated by data mining techniques.

• The results show that CART supplemented by AdaBoost has the best classification performance.

• Time-dependent variables are essential to identify student online learning performance.

摘要

•We develop early warning systems to predict at-risk students while a course is in progress.•Learning portfolios from a fully online course are evaluated by data mining techniques.•The results show that CART supplemented by AdaBoost has the best classification performance.•Time-dependent variables are essential to identify student online learning performance.

论文关键词：Learning management system,e-Learning,Early warning system,Data-mining,Learning performance prediction

论文评审过程：Available online 7 May 2014.

论文官网地址：https://doi.org/10.1016/j.chb.2014.04.002