基于变分自编码器和随机森林的混合式学习风险预警框架

Blended Learning Risk Warning Framework Based on Variational Autoencoder and Random Forest

教育信息化的快速发展,推动了在线学习和混合式学习的普及,学习风险预警研究受到越来越多的关注,然而传统的预警算法都是针对平衡数据集设计,而教育数据集属于高度不平衡数据集,使得这些算法识别风险学生的难度增加。为了解决这个问题,提出一种基于变分自编码器和随机森林的混合式学习风险预警框架(VRFRisk)。框架使用变分自编码器对不平衡数据进行数据平衡处理,利用随机森林算法进行分类。在给定的数据集上的实验结果表明,VRFRisk可以有效处理不平衡的教育数据集,并在召回率和F分数方面提供比基线方法更好、更稳定的预测结果。

The rapid development of educational informationization has promoted the popularization of online learning and blended learning.The research on learning risk early warning has attracted more and more attention.However,the traditional early warning algorithms are designed for balanced datasets,and the education datasets belong to highly unbalanced datasets,which makes it more difficult for these algorithms to identify risk students.To solve this problem,a blended learning risk early warning framework(VRFRisk)based on variational autoencoder and random forest was proposed.The VRFRisk framework used the variational autoencoder to balance the unbalanced data,and used the random forest algorithm to classify.The experimental results on a given data set showed that VRFRisk can effectively handle the imbalanced education dataset and provide much better and more stable prediction results than baseline methods in terms of accuracy and F score.