Curriculum Knowledge Graph Driven Learning Performance Tracking and Adaptive Evaluation Model

Curriculum Knowledge Graph Driven Learning Performance Tracking and Adaptive Evaluation Model This repository provides an implementation of the NodeStatePredictive framework, a curriculum knowledge graph-driven model for learning performance tracking and adaptive evaluation. The model leverages curriculum knowledge graphs to represent relationships among knowledge units and tracks learner states over time. It integrates three key modules: Node Interaction Projector, State Consolidation Module, and Predictive Coding Integrator. These modules jointly model node interactions, consolidate learner states, and predict future learning outcomes. :contentReference[oaicite:0]{index=0} Features - Curriculum knowledge graph-based learner modeling- Node interaction projection for concept relationship representation- Dynamic state aggregation and transition- Temporal predictive coding for future learning state forecasting- Adaptive evaluation for personalized learning support- Experimental support for accuracy, recall, F1-score, AUC, MAE, and transition accuracy Model Overview The framework is built around the NodeStatePredictive model, which contains: 1. Interaction Projection Mechanism2. State Aggregation and Transition3. Temporal Predictive Integration Together, these components support learning performance tracking and adaptive evaluation through graph-based relational reasoning and temporal dependency modeling. :contentReference[oaicite:1]{index=1} Experimental Results Experiments show that NodeStatePredictive outperforms several baseline methods across curriculum knowledge graph, student learning performance, adaptive evaluation, and learning pathway datasets. :contentReference[oaicite:2]{index=2} Installation git clone https://github.com/your-username/your-repository.gitcd your-repositorypip install -r requirements.txtUsagepython train.py --config configs/default.yamlProject Structure├── configs/├── data/├── models/├── scripts/├── train.py├── evaluate.py├── requirements.txt└── README.mdCitation@article{du2025curriculum, title={Curriculum Knowledge Graph Driven Learning Performance Tracking and Adaptive Evaluation Model}, author={Du, Guobiao and Qiu, Feiyue and He, Tao and Guo, Haidong and Wang, Liping}, year={2025}}