Efficient Knowledge Tracing Leveraging Higher-Order Information in Integrated Graphs

2025/07/24 by Donghee Han, Daehee Kim, Han, Donghee +9
Computer Science · #Advanced Graph Neural Networks #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Intelligent Tutoring Systems and Adaptive Learning #Machine Learning (cs.LG) #Online Learning and Analytics

paper · pdf · doi:10.48550/arxiv.2507.18668

openalex publication_date 2025/07/24 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The rise of online learning has led to the development of various knowledge tracing (KT) methods. However, existing methods have overlooked the problem of increasing computational cost when utilizing large graphs and long learning sequences. To address this issue, we introduce Dual Graph Attention-based Knowledge Tracing (DGAKT), a graph neural network model designed to leverage high-order information from subgraphs representing student-exercise-KC relationships. DGAKT incorporates a subgraph-based approach to enhance computational efficiency. By processing only relevant subgraphs for each target interaction, DGAKT significantly reduces memory and computational requirements compared to full global graph models. Extensive experimental results demonstrate that DGAKT not only outperforms existing KT models but also sets a new standard in resource efficiency, addressing a critical need that has been largely overlooked by prior KT approaches.

Citations

• Data-efficient Fine-tuning for LLM-based Recommendation
• LLaMA: Open and Efficient Foundation Language Models
• Enhancing Deep Knowledge Tracing with Auxiliary Tasks
• DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
• DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
• pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models
• Knowledge Tracing: A Survey
• Knowledge Tracing: A Survey
• Inhomogeneous Social Recommendation with Hypergraph Convolutional Networks
• A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions
• Deep Graph Memory Networks for Forgetting-Robust Knowledge Tracing
• Deep Graph Memory Networks for Forgetting-Robust Knowledge Tracing
• Knowledge-aware Coupled Graph Neural Network for Social Recommendation
• Higher-Order Attribute-Enhancing Heterogeneous Graph Neural Networks
• GIKT: A Graph-based Interaction Model for Knowledge Tracing
• Context-Aware Attentive Knowledge Tracing
• Improving Knowledge Tracing via Pre-training Question Embeddings
• EdNet: A Large-Scale Hierarchical Dataset in Education
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• A Self-Attentive model for Knowledge Tracing
• Inductive Matrix Completion Based on Graph Neural Networks
• Graph Attention Networks
• Continuously Differentiable Exponential Linear Units
• Dynamic Key-Value Memory Networks for Knowledge Tracing
• Deep Knowledge Tracing
• Adam: A Method for Stochastic Optimization

Related

• DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
• DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
• Deep Graph Memory Networks for Forgetting-Robust Knowledge Tracing
• Bi-CLKT: Bi-Graph Contrastive Learning based Knowledge Tracing
• Deep Graph Memory Networks for Forgetting-Robust Knowledge Tracing