Capture-Calibrate-Coach: A Graph-Based Framework for Knowledge Monitoring Estimation and Adaptive Feedback
2026-05-25 • Machine Learning
Machine Learning
AI summaryⓘ
The authors created a new system called the 3C framework to help learners better understand what they know and how well they think they know it. Their approach captures learners' self-reported knowledge, uses a graph neural network to estimate understanding even when learners don’t mention everything, and gives personalized feedback based on different learning patterns. Tests showed their method predicted learners' perceived knowledge accurately and that users appreciated the clear, helpful feedback. This work aims to improve learning by supporting both actual knowledge and learners' self-awareness of it.
metacognitionknowledge monitoringself-regulated learningheterogeneous graphgraph neural networkadaptive learninglatent variable inferencepersonalized feedbackAUC (Area Under Curve)user study
Authors
Gen Li, Li Chen, Cheng Tang, Boxuan Ma, Yuncheng Jiang, Daisuke Deguchi, Takayoshi Yamashita, Atsushi Shimada
Abstract
Effective learning support requires understanding not only what learners know but also how accurately they perceive their own understanding. This metacognitive dimension, known as knowledge monitoring, fundamentally influences self-regulated learning, yet this dimension remains underexplored in current systems. This paper introduces the Capture-Calibrate-Coach (3C) framework for adaptive learning support. The Capture phase extracts learners' perceived knowledge states from open-ended self-reports to construct a heterogeneous graph linking learners and knowledge concepts. The Calibrate phase applies a heterogeneous graph neural network to infer latent perceived states for concepts not explicitly mentioned, enabling systematic knowledge monitoring assessment. The Coach phase classifies learners into five metacognitive patterns and delivers personalized feedback addressing both knowledge gaps and calibration errors. Evaluation with 684 students demonstrates 85.21% AUC in predicting latent perceived states, significantly outperforming baseline methods. A user study with 47 participants shows positive reception of feedback quality, with participants particularly valuing concrete feedback on knowledge gaps and actionable study guidance. These findings advance AI-based learning support toward metacognitive teammates that foster accurate self-awareness while supporting knowledge growth.