Abstract—In the field of online learning, there is a problem of high student turnover rate. How to accurately identify learners and provide targeted teaching support services is an urgent problem for education researchers. In this paper, 1306 online learners majoring in finance from Shanghai Open University were selected as the subjects, and two kinds of data sets are adopted, which are learning data of online learning platform and learning behavior data of students based on xAPI, to analyze the relationship between learners' various online learning behaviors and learning achievements, and to determine the characteristics related to learning state of learners, describe the personalized learning state portrait, and select a variety of machine learning algorithms to build prediction model based on two data sets, to explore which data is more effective for building prediction models to identify potential risk learners. It is found that data mining analysis based on xAPI data has higher prediction accuracy than traditional online learning data.
Index Terms—Data mining, xAPI, adaptive, learning prediction.
Jun Xiao and Lamei Wang are with the Shanghai Engineering Research Center of Open Distance Education, Shanghai Open University, China (e-mail: ecnuxj2003@163.com, wlmyisa@163.com). Jisheng Zhao is with the Foda intelligence Inc., China (e-mail: zhaojisheng@foda.ai). Aizhen Fu is with the East China Normal University, China (e-mail: aienfu@163.com).
Cite:Jun Xiao, Lamei Wang, Jisheng Zhao, and Aizhen Fu, "Research on Adaptive Learning Prediction Based on XAPI," International Journal of Information and Education Technology vol. 10, no. 9, pp. 679-684, 2020.
Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).