HISE-KT: Synergizing Heterogeneous Information Networks and LLMs for Explainable Knowledge Tracing with Meta-Path Optimization

2025-11-19

AI Summary

Knowledge tracing (KT) faces two key challenges: (1) HIN-based approaches suffer from noise due to manual or random meta-path selection and lack quality assessment of meta-path instances; (2) LLM-based methods neglect inter-student relational modeling, hindering evidence-grounded, interpretable predictions. To address these, we propose a unified HIN-LLM fusion framework: (i) we construct a multi-relational graph encoding student–item–concept interactions; (ii) we design an educational-psychology-inspired similar-student retrieval mechanism; (iii) we introduce the first LLM-driven semantic scoring and automated filtering of meta-path instances for principled meta-path optimization; and (iv) we employ structured prompting to generate interpretable, evidence-supported analytical reports. Evaluated on four public KT benchmarks, our method achieves statistically significant improvements in both prediction accuracy and explanation fidelity. It is the first to enable evidence-based, personalized learning analytics grounded in both structural relational knowledge and generative reasoning.

Technology Category

• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Machine Learning: Statistical Relational/Logic Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Knowledge Tracing (KT) aims to mine students'evolving knowledge states and predict their future question-answering performance. Existing methods based on heterogeneous information networks (HINs) are prone to introducing noises due to manual or random selection of meta-paths and lack necessary quality assessment of meta-path instances. Conversely, recent large language models (LLMs)-based methods ignore the rich information across students, and both paradigms struggle to deliver consistently accurate and evidence-based explanations. To address these issues, we propose an innovative framework, HIN-LLM Synergistic Enhanced Knowledge Tracing (HISE-KT), which seamlessly integrates HINs with LLMs. HISE-KT first builds a multi-relationship HIN containing diverse node types to capture the structural relations through multiple meta-paths. The LLM is then employed to intelligently score and filter meta-path instances and retain high-quality paths, pioneering automated meta-path quality assessment. Inspired by educational psychology principles, a similar student retrieval mechanism based on meta-paths is designed to provide a more valuable context for prediction. Finally, HISE-KT uses a structured prompt to integrate the target student's history with the retrieved similar trajectories, enabling the LLM to generate not only accurate predictions but also evidence-backed, explainable analysis reports. Experiments on four public datasets show that HISE-KT outperforms existing KT baselines in both prediction performance and interpretability.

Problem

• Optimizes meta-path selection to reduce noise in knowledge tracing models
• Integrates heterogeneous networks with LLMs for explainable student performance predictions
• Automates quality assessment of meta-paths and retrieves similar student learning patterns

Innovation

• Integrates heterogeneous networks with large language models
• Automates meta-path quality assessment using LLM scoring
• Retrieves similar student trajectories for explainable predictions