Citation Author(s):
East China Normal University
Submitted by:
Yuang Wei
Last updated:
Mon, 06/03/2024 - 03:28
0 ratings - Please login to submit your rating.


IRIC method's data and code are available at this URL.

These data links contain publicly available datasets that can be downloaded directly from their website. Our research on IRIC has validated the performance of the model through these publicly available datasets. Please continue to pay attention.

These data mainly include Emergency Event Data (ALARM) and Education Dataset (Junyi), which can be used for research in causal structure learning, knowledge tracking, and other areas.

The abstract of our IRIC method are:
The discovery of relationships among knowledge components benefits personalized learning path generation and resource recommendation in adaptive learning systems. The commonly used correlation-based approaches fail to extract true causal relationships among knowledge components. Moreover, the traditional Bayesian network structure learning methods are overly generalized, missing potential insights during the learning process. To address this issue, this work proposes a causal structure learning method via Item Response Theory (IRT), which uncovers latent relationships among knowledge components by integrating information entropy and causal effects. Furthermore, a structure search algorithm based on multi-population co-evolutionary optimization is proposed to enhance the efficiency and accuracy of structural search. The proposed approach outperformed several existing algorithms on both public datasets and real-world datasets. This study offers an automated method for obtaining knowledge systems and structures in adaptive learning systems, laying the groundwork for further studies in cognitive diagnosis, resource recommendation, and learning path planning in personalized learning research. Our program code is accessible at


IRIC method's data and code are available at this URL.