HCGKT: Hierarchical Contrastive Graph Knowledge Tracing with Multi-level Feature Learning

We added HCGKT into our pyKT package.

The link is here and the API is here.

Original paper can be found at ZhiW,Huang et al. “HCGKT: Hierarchical Contrastive Graph Knowledge Tracing with Multi-level Feature Learning.” International Conference on Artificial Intelligence in Education. 2025.

Title: HCGKT: Hierarchical Contrastive Graph Knowledge Tracing with Multi-level Feature Learning

Abstract: Knowledge Tracing (KT) aims to predict learners’ future performance by modeling their historical interaction data. Despite recent advances in attention-based KT models, significant challenges remain: effectively capturing hierarchical relationships between questions and knowledge components (KCs), handling noisy educational data, and accurately modeling complex semantic relationships. To address these challenges, we propose Hierarchical Contrastive Graph Knowledge Tracing (HCGKT), which combines hierarchical graph filtering attention, adversarial contrastive learning, and graph convolutional networks. Experiments on three datasets demonstrate our model’s superior performance in both prediction accuracy and interpretability. We have provided all the datasets and code at https://pykt.org/.