%0 Conference Paper %1 Zhang_2023 %A Zhang, Moyu %A Zhu, Xinning %A Zhang, Chunhong %A Qian, Wenchen %A Pan, Feng %A Zhao, Hui %B Proceedings of the 32nd ACM International Conference on Information and Knowledge Management %D 2023 %I ACM %K cikm2023 knowledge-tracing %P 3236-3246 %R 10.1145/3583780.3614827 %T Counterfactual Monotonic Knowledge Tracing for Assessing Students\textquotesingle Dynamic Mastery of Knowledge Concepts %U https://doi.org/10.1145%2F3583780.3614827 %X As the core of the Knowledge Tracking (KT) task, assessing students' dynamic mastery of knowledge concepts is crucial for both offline teaching and online educational applications. Since students' mastery of knowledge concepts is often unlabeled, existing KT methods focus on predicting students' responses to practices. However, purely predicting student responses without imposing specific constraints on hidden concept mastery values does not guarantee the accuracy of these intermediate values as concept mastery values. To address this issue, we propose a principled approach called Counterfactual Monotonic Knowledge Tracing (CMKT), which builds on the implicit paradigm described above by using a counterfactual assumption to constrain the evolution of students' mastery of knowledge concepts. Specifically, CMKT first assesses students' knowledge concept mastery value based on their historical practice sequences. Then, CMKT sets the answer of the most recent practice as the opposite of the actual answer and, based on this counterfactual answer, assesses the student's corresponding counterfactual knowledge mastery value. During the model training process, CMKT constrains the update of the student's knowledge states by ensuring that the two types of knowledge mastery values of students satisfy a fundamental educational theory, the monotonicity theory, to provide specific semantics for the assessed mastery values by the model. Finally, extensive experiments on five datasets demonstrate the superiority of CMKT over baseline models.

@inproceedings{Zhang_2023, abstract = {As the core of the Knowledge Tracking (KT) task, assessing students' dynamic mastery of knowledge concepts is crucial for both offline teaching and online educational applications. Since students' mastery of knowledge concepts is often unlabeled, existing KT methods focus on predicting students' responses to practices. However, purely predicting student responses without imposing specific constraints on hidden concept mastery values does not guarantee the accuracy of these intermediate values as concept mastery values. To address this issue, we propose a principled approach called Counterfactual Monotonic Knowledge Tracing (CMKT), which builds on the implicit paradigm described above by using a counterfactual assumption to constrain the evolution of students' mastery of knowledge concepts. Specifically, CMKT first assesses students' knowledge concept mastery value based on their historical practice sequences. Then, CMKT sets the answer of the most recent practice as the opposite of the actual answer and, based on this counterfactual answer, assesses the student's corresponding counterfactual knowledge mastery value. During the model training process, CMKT constrains the update of the student's knowledge states by ensuring that the two types of knowledge mastery values of students satisfy a fundamental educational theory, the monotonicity theory, to provide specific semantics for the assessed mastery values by the model. Finally, extensive experiments on five datasets demonstrate the superiority of CMKT over baseline models. }, added-at = {2023-10-24T17:07:11.000+0200}, author = {Zhang, Moyu and Zhu, Xinning and Zhang, Chunhong and Qian, Wenchen and Pan, Feng and Zhao, Hui}, biburl = {https://www.bibsonomy.org/bibtex/2b649ea201bcadec1aa74623e489975ae/brusilovsky}, booktitle = {Proceedings of the 32nd {ACM} International Conference on Information and Knowledge Management}, description = {Counterfactual Monotonic Knowledge Tracing for Assessing Students' Dynamic Mastery of Knowledge Concepts | Proceedings of the 32nd ACM International Conference on Information and Knowledge Management}, doi = {10.1145/3583780.3614827}, interhash = {f61cecb8b7ad9c5fd63be4457419e221}, intrahash = {b649ea201bcadec1aa74623e489975ae}, keywords = {cikm2023 knowledge-tracing}, month = oct, pages = {3236-3246}, publisher = {{ACM}}, timestamp = {2023-10-24T17:07:11.000+0200}, title = {Counterfactual Monotonic Knowledge Tracing for Assessing Students{\textquotesingle} Dynamic Mastery of Knowledge Concepts}, url = {https://doi.org/10.1145%2F3583780.3614827}, year = 2023 }