Learning Management System Log Data Analysis via Machine Learning

• In Event: Technical Issues in Data Mining and Learning Analytics

Abstract

Previous studies analyzing learning management system (LMS) log data employed traditional statistical methods. This study employed penalized regression (group enet and group mnet) among machine learning methods to analyze small but wide LMS log data from a graduate-level flipped classroom. As results, penalized regression not only produced prediction models comparable to random forest (a method well-known for its prediction capacity) in terms of prediction but also identified important variables to predict students’ academic achievement. In particular, this study was able to predict low-performing students as early as the first 2 to 3 weeks of the course without quiz or exam. Scientific importance of the research was discussed.

Authors

• Jin Eun Yoo, Korea National University of Education

• Minjeong Rho, Korea National University of Education