MOOC中学生流失现象分析与预警

he campaign of Massive Open Online Courses (MOOCs) in the area of e-learning and distant education gains significant popularity among both students and educators.As observation of MOOC gets deeper, studies pointed out that student, however, can hardly finish a course on MOOCs. Less than average 5% of students could follow course. Such a low retention rate will restrict development of MOOCs in future. This paper shows observation of data on MOOCs. A factor analysis is applied to students' behavioral data. The result is discussed with course design on MOOC platform. A general online system for predicting students' churn is developed. The system consists of sampler, classifier, optional differentiator and optional attenuato. With unsupervised learning, this system can fit on different on-going courses. A discussion of this system with the course Data Structures and Algorithms is conducted. To test the scalability of the system, the system also applied on the course Introduction to Computing. Scalability of the system is fine but an analysis of course and to choose optional components based on the analysis is required. Additionally, based on research above, some suggestions for instructing students on MOOC is given.