Personalized Exercise Recommendation with Semantically-Grounded Knowledge Tracing

2025-07-15

AI Summary

Existing exercise recommendation methods predominantly rely on knowledge tracing (KT) while neglecting item semantics and the sequential, structured evolution of student learning. To address this, we propose ExRec, an end-to-end personalized exercise recommendation framework that jointly integrates semantic representation learning, structure-aware sequential KT modeling, and model-augmented Q-learning. Specifically, ExRec employs a semantic encoder to capture item-level conceptual meaning; designs a structure-aware sequential KT module to explicitly model the hierarchical and temporal progression of learning states; and introduces a model-based value estimation mechanism to enhance generalization to unseen exercises and improve interpretability of recommended learning paths. Evaluated on four real-world online mathematics learning datasets, ExRec consistently outperforms state-of-the-art baselines across multiple metrics. It supports cross-task transfer learning and generates cognitively grounded, interpretable learning trajectories—demonstrating both efficacy and pedagogical validity.

Technology Category

• Machine Learning: Structured Learning
• Data Mining & Knowledge Management: Recommender Systems
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

We introduce ExRec, a general framework for personalized exercise recommendation with semantically-grounded knowledge tracing. Our method builds on the observation that existing exercise recommendation approaches simulate student performance via knowledge tracing (KT) but they often overlook two key aspects: (a) the semantic content of questions and (b) the sequential, structured progression of student learning. To address this, our ExRec presents an end-to-end pipeline, from annotating the KCs of questions and learning their semantic representations to training KT models and optimizing several reinforcement learning (RL) methods. Moreover, we improve standard Q-learning-based continuous RL methods via a tailored model-based value estimation (MVE) approach that directly leverages the components of KT model in estimating cumulative knowledge improvement. We validate the effectiveness of our ExRec using various RL methods across four real-world tasks with different educational goals in online math learning. We further show that ExRec generalizes robustly to new, unseen questions and that it produces interpretable student learning trajectories. Together, our findings highlight the promise of KT-guided RL for effective personalization in education.

Problem

• Existing methods ignore question semantics in exercise recommendation
• Current approaches neglect structured learning progression in knowledge tracing
• Standard RL lacks KT-integrated value estimation for knowledge improvement

Innovation

• Semantically-grounded knowledge tracing for exercise recommendation
• End-to-end pipeline with reinforcement learning optimization
• Model-based value estimation leveraging KT components