Read the full-text of this article for free

Title: Tracking domain knowledge in Chinese language education based on graph neural networks

Authors: Wenjuan Hu; Jing Fan; Yan Wang

Addresses: Hebei Institute of Mechanical and Electrical Technology, Xingtai, 054000, China ' Hebei Institute of Mechanical and Electrical Technology, Xingtai, 054000, China ' Hebei Institute of Mechanical and Electrical Technology, Xingtai, 054000, China

Abstract: Knowledge tracking (KT) is a core task in the domain of integrated Chinese language education. However, traditional KT methods struggle to fully uncover the complex knowledge relationships within Chinese language education. To address this, this article designs a knowledge heterogeneous graph in the domain of Chinese language education, designs a heterogeneous graph neural network (GNN) to learn interactive relations among nodes, and extracts exercise node features as exercise representations. Then, a deep residual network is suggested to learn the interaction among exercise representations and students' answering abilities. Finally, a temporal convolutional network is used to track students' cognitive states and forecast the probability of them correctly answering the next exercise. Experimental results on the ASSIST and KDD datasets show that the proposed method improves prediction accuracy by at least 2.74% and 3.41%, respectively, enabling accurate forecasting of the mastery level of Chinese language knowledge points.

Keywords: Chinese language education integration field; knowledge tracking; graph neural network; GNN; deep residual network; temporal convolutional network; TCN.

DOI: 10.1504/IJICT.2025.151078

International Journal of Information and Communication Technology, 2025 Vol.26 No.51, pp.34 - 49

Received: 24 Aug 2025
Accepted: 13 Nov 2025
Published online: 12 Jan 2026 *