Meta Transfer Learning for Early Success Prediction in MOOCs

TL;DR

This paper introduces three novel transfer strategies for early success prediction in MOOCs, enabling models trained on diverse courses to predict student outcomes across different domains effectively.

Contribution

The paper proposes three innovative transfer learning strategies—pre-training on multiple courses, incorporating course meta information, and fine-tuning on previous course iterations—to improve early success prediction in MOOCs.

Findings

Models with interaction data and course info outperform previous iteration models.

Transfer learning strategies achieve comparable or better performance.

Effective early prediction models can be developed with limited course-specific data.

Abstract

Despite the increasing popularity of massive open online courses (MOOCs), many suffer from high dropout and low success rates. Early prediction of student success for targeted intervention is therefore essential to ensure no student is left behind in a course. There exists a large body of research in success prediction for MOOCs, focusing mainly on training models from scratch for individual courses. This setting is impractical in early success prediction as the performance of a student is only known at the end of the course. In this paper, we aim to create early success prediction models that can be transferred between MOOCs from different domains and topics. To do so, we present three novel strategies for transfer: 1) pre-training a model on a large set of diverse courses, 2) leveraging the pre-trained model by including meta information about courses, and 3) fine-tuning the model on previous course iterations. Our experiments on 26 MOOCs with over 145,000 combined enrollments and millions of interactions show that models combining interaction data and course information have comparable or better performance than models which have access to previous iterations of the course. With these models, we aim to effectively enable educators to warm-start their predictions for new and ongoing courses.