Deep Attentive Study Session Dropout Prediction in Mobile Learning Environments
"World’s first study to predict when users will stop studying in a mobile environment"
Accepted as a full paper at International Conference on Computer Supported Education (CSEDU) 2020
Analyzing a relatively short session-based mobile learning environment, Riiid developed a deep learning Transformer-based predictive model called Deep Attentive Study Session Description (DAS). The model defines the concept of user ‘dropout’ for the first time and accurately predicts dropout rates by examining various learning behaviors of users.
The DAS-based prediction model outperforms the best LSTM and GRU models by 12.2 points based on Area Under the receiver operating characteristic Curve (AUC).