Student proficiency prediction on CNMOOC data

Abstract

MOOC continues to thrive in today and CNMOOC is one of the largest MOOC platform in China which cooperates with many high schools and universities. We try to hardness the data of students' learning behaviors to provide better personalized learning advice. Online education digitalizes students' learning behaviors which makes it convenient for analyzing students' behaviors. However the industrial data of MOOC is much sophisticated and sparse. In the paper, we introduce the system using machine learning methods to improve educational outcomes and describe some ideas tackling data sparsity in the scenario. We only focus on predicting student learning proficiency on specific course and compare different models under the scenario of inadequate data.