CLMS-Net:Dropout Prediction in MOOCs with Deep Learning
Massive Open Online Courses(MOOCs)have played an increasingly crucial role in education,but the high dropout rate problem is great serious.Predicting whether students will dropout has attracted considerable attention.Current methods mainly depend on handcrafted features.The process is laborious and not scalability,and even difficult to guarantee the final prediction effect.In this paper,we propose a deep neural network model,which is a combination of Convolutional Neural Network,Long Short-Term Memory network and Support Vector Machine.Our model has an effective feature extraction strategy,which automatically extract features from the raw data,and takes into account the impact of the sequential relationship of student behavior and class imbalance on dropout and,most importantly,reinforce the performance of dropout prediction.Extensive experiments on a public dataset have shown that the proposed model can achieve better results comparable to feature engineering based methods and other neural network methods.
MOOCs Dropout prediction Deep learning Learning analytics
Nannan Wu Lei Zhang Yi Gao Mingfei Zhang Xia Sun Jun Feng
College of Information Science and Technology Northwest University Xi'an,China
国际会议
2019国图灵大会(ACM Turing Celebration conference-China 2019 )
成都
英文
99-104
2019-05-17（万方平台首次上网日期，不代表论文的发表时间）