Deep-Sequence–Aware Candidate Generation for e-Learning System

Recently proposed recommendation systems based on embedding vector technology allow us to utilize a wide range of information such as user side and item predict preferences. Since there is lack ability use the sequential history, most system algorithms fail user’s preferences more accurately. Therefore, in this study, we developed novel that takes advantage sequence heterogeneous candidate-generation process. The principle underlying model new layer catches pattern history. deep-learning may improve prediction accuracy using data, profile. Experiments were conducted datasets Korean e-learning platform, empirical results confirmed capability approach its superiority over models do not sequences users items for