Systematic ensemble model selection approach for educational data mining

作者：

Highlights：

• Implement EDM techniques to identify at risk students during course delivery.

• Analyze two educational datasets at two course delivery stages using various methods.

• Propose a systematic approach based on Gini index and p-value for ensemble selection.

• Results show that the ensemble models achieve high accuracy and high specificity.

摘要

•Implement EDM techniques to identify at risk students during course delivery.•Analyze two educational datasets at two course delivery stages using various methods.•Propose a systematic approach based on Gini index and p-value for ensemble selection.•Results show that the ensemble models achieve high accuracy and high specificity.

论文关键词：e-learning,Student performance prediction,Educational data mining,Ensemble learning model selection,Gini index,p-value

论文评审过程：Received 27 November 2019, Revised 1 April 2020, Accepted 29 April 2020, Available online 7 May 2020, Version of Record 16 May 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.105992