The massive open online course (MOOC) has expanded rapidly, providing users with a low-cost, high-quality learning experience. High dropout rate is a serious obstacle that restricts the development of ideological and political MOOC. One of the ways to solve this obstacle is to use the rich data resources in MOOC to explore the relevant factors of dropout. Reduce dropout rates by building drop-out prediction models and establishing early-warning mechanisms. However, the ideological MOOC data is huge and complex, which is prone to problems such as loss of data value, mismatch between data and models, and poor research reproducibility. This paper uses a more mature logistic regression method of machine learning to transfer it to the field of education, providing a new path for data-driven MOOC dropout prediction research.
Zhang, Yan; Zhang, Qian; and Liu, Xu, "Dropout Predictions of Ideological and Political MOOC Learners Based on Big Data" (2020). WHICEB 2020 Proceedings. 30.