Time-and-Concept Enhanced Deep Multidimensional Item Response Theory for interpretable Knowledge Tracing

作者：

Highlights：

• An improved knowledge tracing model that integrates multi-dimensional IRT into deep networks.

• Exploit temporal signal through the zoom gate.

• Interpretable parameters show examinee and item properties directly.

摘要

•An improved knowledge tracing model that integrates multi-dimensional IRT into deep networks.•Exploit temporal signal through the zoom gate.•Interpretable parameters show examinee and item properties directly.

论文关键词：Knowledge Tracing,Recurrent neural network,Interpretable machine learning,Item Response Theory

论文评审过程：Received 3 August 2020, Revised 25 January 2021, Accepted 30 January 2021, Available online 15 February 2021, Version of Record 26 February 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.106819