EDPS: Early Dropout Prediction System of MOOC Courses

In the last few years, Massive open online courses (MOOCs) have become the major online learning method worldwide. However, the high dropout rate has severely hampered its development. But most of current machine learning, deep learning, and ensemble learning methods have deficiency in learning effective features. Therefore, this paper has proposed an early dropout prediction method, and developed a system for the personalized education guidance of students and teachers. Attention-based document representation as a vector (A-Doc2vec) is proposed to learn sequence features of course and video, and heterogeneous classification model is proposed to improve the dropout prediction accuracy. This system has improved the students’ commitment to their studies, and reduced the dropout rate of MOOC Courses.