Online Learning Behavior Analysis and Achievement Prediction with Explainable Machine Learning

Abstract

With the development of computer technology, “Internet + education" has become a hot topic, which has promoted the development of online education. The impact of the epidemic has brought new challenges to online education, and the effect of online teaching has sparked extensive research. Why online learning has not received satisfying learning effect? And how to improve the quality of online education have become the focus of research. To address these challenges in online learning, this paper studies the important factors affect the students’ learning achievements based on the explicit and implicit features extracted from the online learning behavior data. And a wealth of machine learning models are explored to predict the learner’s achievement, including the original linear models, the ensemble learning models and the advanced deep learning models. With extensive experimental results achieved, we gain reasonable explanation on the influence factors hander the learning achievement, and predict the learning achievements with the explainable machine learning models. Finally, some suggestions are put forward to improve students’ learning behaviors, aiming to promote the quality of online learning in the long run.