Massive Open Online Courses (MOOC) became popular and they posted great impact to education. Students could enroll and attend any MOOC anytime and anywhere according to their interest, schedule and learning pace. However, the dropout rate of MOOC was known to be very high in practice. It is desirable to discover students who have high chance to dropout in MOOC in early stage, and the course leader could impose intervention immediately in order to reduce the dropout rate. In this paper, we proposed a framework that applies big data methods to identify the students who are likely to dropout in MOOC. Real- world data were collected for the evaluation of our proposed framework. The results demonstrated that our framework is effective and helpful. Copyright © 2015 Springer-Verlag Berlin Heidelberg.
|Title of host publication||Technology in education. Technology-mediated proactive learning: Second international conference, ICTE 2015, Hong Kong, China, July 2-4, 2015, Revised Selected Papers|
|Editors||Jeanne LAM, Kwan Keung NG, Simon K. S. CHEUNG, Tak Lam WONG, Kam Cheong LI, Fu Lee WANG|
|Place of Publication||Germany|
|Publisher||Springer Berlin Heidelberg|
|Publication status||Published - 2015|
CitationTang, J. K. T., Xie, H., & Wong, T.-L. (2015). A big data framework for early identification of dropout students in MOOC. In J. Lam, K. K. Ng, S. K. S. Cheung, T. L. Wong, K. C. Li, & F. L. Wang (Eds.), Technology in education. Technology-mediated proactive learning: Second international conference, ICTE 2015, Hong Kong, China, July 2-4, 2015, Revised Selected Papers (pp. 127-132). Germany: Springer Berlin Heidelberg.
- Big data
- Decision tree
- Dropout rate