Predicting High-Risk Students Using Learning Behavior

Over the past few years, growing popularity of online education has enabled there to be a large amount students’ learning behavior data stored, which brings great opportunities and challenges field educational mining. Students’ performance can predicted, based on data, so as identify at-risk students who need timely help complete their studies improve teaching quality. In order make full use these new prediction method was designed existing research. This constructs hybrid deep model, simultaneously obtain temporal information overall from that it more accurately predict high-risk students. When compared with methods, experimental results show proposed offers better predicting performance.