Deep Knowledge Tracing Based on Behaviour in the Item Response Theory Framework

• Author / Creator

  Ma, Chaojun

• Knowledge tracing involves assessing students' performance based on their learning interaction records and providing them with personalized learning paths. Deep learning methods model students' knowledge states by exploring extensive exercise records, with the Deep-IRT approach considered superior to others. However, Deep-IRT overlooks relevant behavioural features that could assist in modelling students' knowledge states. Therefore, this study introduces a new method for modelling student behaviour, combining behavioural features with learning records to enhance knowledge tracking performance. Experimental results on two real-world benchmark datasets indicate that the proposed model significantly outperforms the Deep-IRT model in predicting future students' abilities. Limitations of the current study and potential directions for future research are also discussed.

• Subjects / Keywords

  • Knowledge Tracing
  • Deep learning

• Graduation date

  Spring 2024

• Type of Item

  Thesis

• Degree

  Master of Education

• DOI

  https://doi.org/10.7939/r3-s4jq-m449

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.