austin2029/Burmese-nlp-dkt

Research Paper & Overview

This repository hosts the official implementation and model weights for the research paper:
"Integrating Natural Language Processing with Deep Knowledge Tracing in Intelligent Tutoring Systems for Low-Resource Languages: A Case Study on Burmese STEM Curricula"[cite: 8]

• Author: Aung Ko Ko Oo (Burmese Artificial Intelligence Research Institute & University of the People)[cite: 8]
• Preprint / DOI: Research Square (DOI: 10.21203/rs.3.rs-9991485/v1)[cite: 8]

---

Evaluation Results

Model Variant	AUC	RMSE
Baseline Text-Blind Binary DKT[cite: 8]	0.762[cite: 8]	0.412[cite: 8]
Full Hybrid NLP-DKT (Proposed)[cite: 8]	0.824[cite: 8]	0.345[cite: 8]

---

Citation

If you use this model, dataset, or code in your academic research, please cite the paper as follows:[cite: 9]

@article{oo2026hybridnlpdtk,
  title={Integrating Natural Language Processing with Deep Knowledge Tracing in Intelligent Tutoring Systems for Low-Resource Languages: A Case Study on Burmese STEM Curricula},
  author={Aung Ko Ko Oo},
  journal={Burmese Artificial Intelligence Research Institute & University of the People},
  year={2026}
}