Knowledge Tracing (KT) refers to inferring the students' knowledge mastery and predicting their future performance. KT serves as the foundation for personalized learning and enhances the effectiveness of educational interventions, becoming a crucial technology in intelligent tutoring systems. Recent approaches have demonstrated notable success by harnessing the potent representational capacities of deep learning. However, complex neural networks lead to entangled knowledge state embeddings, where the embedding dimensions are coupled, limiting their expressiveness and interpretability. In addition, the limitations of existing methods in Euclidean space result in distortions when capturing complex relationships among knowledge states. This distortion manifests as the distance and geometric structures among knowledge states being deformed during the embedding process. To address the challenges, in this paper, we propose a hyperbolic hypergraph transformer with knowledge stateDisentanglement forKnowledgeTracing, named DisenKT. We construct the students' response sequences into the hypergraph, projected into the hyperbolic space to alleviate the representation distortion problem of questions and knowledge states. The embeddings of hierarchical knowledge states are refined through message passing between questions and students based on the proposed hyperbolic hypergraph transformer. Moreover, we are the first to disentangle knowledge states via a contrastive clustering auxiliary task, which enhances the expressiveness and interpretability of knowledge state embeddings. Extensive experimental results on three public datasets demonstrate that DisenKT outperforms state-of-the-art methods on student performance prediction and interpretability.

Knowledge Tracing (KT) aims to assess students’ mastery of knowledge concepts and predict their performance from their historical response records. However, most existing KT models only consider the correspondence between questions and knowledge concepts given in the dataset when constructing the question-knowledge concept structure. This approach fails to deeply explore the knowledge concepts hidden in the questions and may propagate mislabeled associations. In addition, they overlook the impact of individual differences in students’ learning abilities on the accuracy of KT models. In this paper, we propose EGANKT to solve the above problems by introducing a knowledge concept prediction module and a learning ability prediction module. The knowledge concept prediction module identifies and optimizes potential knowledge concepts within questions, mitigating mislabeling issues. The learning ability prediction module improves the prediction accuracy of the model by calculating students’ learning abilities. To further improve the accurate modeling of knowledge states, we introduce self-supervised tasks to support the KT task through data augmentation and contrastive learning of knowledge states. Experimental results show that the EGANKT model outperforms baseline models on five datasets, demonstrating the effectiveness of our approach.