/HAKT

Hierarchical Attention Network based Explainable Knowledge Tracing

Primary LanguagePython

HAKT

This repository contains the source code of our paper: 《基于多层注意力网络的可解释认知追踪方法》(Hierarchical Attention Network based Interpretable Knowledge Tracing).

Paper download link: https://crad.ict.ac.cn/CN/10.7544/issn1000-1239.2021.20210997

update

We upload the preprocessed ASSIST09 dataset as data.zip (2022-04-04).

We upload the "Package requirements file" as requirements.txt (2022-5-18)

Data Preparation

You can download original data form the following links: --

ASSIST09: https://sites.google.com/site/assistmentsdata/home/2009-2010-assistment-data

ASSIST12: https://sites.google.com/site/assistmentsdata/2012-13-school-data-with-affect

ASSIST17: https://sites.google.com/view/assistmentsdatamining/dataset

EdNet-KT1: https://github.com/riiid/ednet

Statics2011: https://pslcdatashop.web.cmu.edu/DatasetInfo?datasetId=507

Eedi: https://eedi.com/projects/neurips-education-challenge

Cite

If you find this code useful in your research then please consider citing the following paper:

@article{孙建文2021基于多层注意力网络的可解释认知追踪方法,
  title={基于多层注意力网络的可解释认知追踪方法},
  author={孙建文 and 周建鹏 and others},
  journal={计算机研究与发展},
  volume={58},
  number={12},
  pages={2630},
  year={2021}
}

If you need more information about our experiments, you can contact us via email: hi.zhou@mails.ccnu.edu.cn