Modeling learning behaviors and predicting performance in an intelligent tutoring system: a two-layer hidden Markov modeling approach

ABSTRACT

To better understand the self-regulated learning process in online learning environments, this research applied a data mining method, the two-layer hidden Markov model (TL-HMM), to explore the patterns of learning activities. We analyzed 25,818 entries of behavior log data from an intelligent tutoring system. Results indicated that students with different learning outcomes demonstrated distinct learning patterns. Students who failed a problem set exhibited more passive learning behaviors and could hardly learn from practice, while students who mastered a problem set could effectively regulate their learning. Furthermore, we extended the use of TL-HMM to predicting learning outcome from behavior sequences and checked through cross-validation. TL-HMM is demonstrated helpful to gain insight into learners’ interactions with online learning environments. In practice, TL-HMM could be embedded in intelligent tutoring systems to monitor learning behaviors and learner status, so as to detect the difficulties of learners and facilitate learning.

KEYWORDS:

• The two-layer hidden Markov model
• self-regulated learning
• learning prediction
• intelligent tutoring system
• educational data mining

Acknowledgements

The authors thank McGraw-Hill Education for providing the data and thank Learnta Inc. for financial and technical support. The results and opinions expressed in this article are the authors’ own and do not reflect the views of McGraw-Hill Education or Learnta.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data are not to be shared publicly due to commercial restrictions, so supporting data is not available.

Additional information

Funding

This work was financially supported by the National Natural Science Foundation of China [#61807011] and the self-determined research funds of CCNU from the colleges’ basic research and operation of MOE [#CCNU19TD019] granted to YT, and by the Institute of Education Sciences of U.S. Department of Education [#R305A090528] granted to XH.

Notes on contributors

Yun Tang

Yun Tang is an associate professor at School of Psychology, Central China Normal University and affiliated with the Key Laboratory of Adolescent Cyberpsychology and Behavior, Ministry of Education. Her research interests include self-regulated learning in the online learning environments, educational data mining, and the self-determination theory of motivation.

Zhengfan Li

Zhengfan Li is currently a graduate student at School of Psychology, Central China Normal University. Her research interests are self-regulation learning and online learning.

Guoyi Wang

Guoyi Wang is currently a graduate student at College of Engineering, North Carolina State University. Her research interests include educational data mining and website development.

Xiangen Hu

Xiangen Hu is a professor and Dean of School of Psychology, Central China Normal University, and a professor at Department of Psychology, University of Memphis. His primary research areas include Mathematical Psychology, Research Design and Statistics, and Cognitive Psychology. More specific research interests include General Processing Tree (GPT) models, categorical data analysis, knowledge representation, computerized tutoring, and advanced distributed learning.