Natural Language Processing

This project is to do SID by RoBERTa-CNN on the online corpus dataset [ArXiv]. 

I guide students to do the entire research. The paper got accepted by IEEE EMBC 2024.

We utilized Generative Pre-trained Transformer (GPT) API to clean up the dataset.

CONCLUSION: The proposed automatic SID system achieved 98% accuracy compared to the state-of-the-art methods.