A Visual Analytic Approach for Presenting Learning Trajectories Along With the Mastery Level of Knowledge

• In Event: Tracking and Detracking Roundtable

Abstract

Visualizing the student sequential learning behaviors and estimating the knowledge gaining at each timestep, has significant implications for updating the learning strategies to improve the learning performance in online learning settings. To this end, a specific online learning analytics algorithm will be proposed, based on data visualization and machine learning algorithms. In particular, this approach can be used to analyze and visualize different learning behavior sequences and provide information for clustering students into different learning groups. Moreover, it can also estimate students’ knowledge mastery level at each timestep based on Recurrent Neural Network (RNN) algorithms using the historical learning behaviors for students. The proposed models will be tested on the learning data of 3,987 students from Algebra Nation in Florida.

Authors

• bo pei, University of Florida

• Wanli Xing, University of Florida