BiomedBERT-fulltext_finetuned_DiMB-RE_RE

Fine-tuned RE Model for DiMB-RE

Model Description

This is a fine-tuned Relation Extraction (RE) model based on the BiomedNLP-BiomedBERT-base-uncased model, specifically designed for sentence classification task to extract relations between extract entities for diet, human metabolism and microbiome field. The model has been trained on the DiMB-RE dataset and is optimized to infer relationship with 13 relation types.

<!-- ### Key Features: - **Language**: English - **Task**: Token classification for Named Entity Recognition (NER) - **Base Model**: BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext - **Domains**: Biomedical, Clinical, Scientific -->

Performance

The model has been evaluated on the DiMB-RE using the following metrics:

• RE (w/ GOLD entities and triggers) - P: 0.799, R: 0.772, F1: 0.785
• RE (Strict, w/ predicted entities and triggers) - P: 0.416, R: 0.336, F1: 0.371
• RE (Relaxed, w/ predicted entities and triggers) - P: 0.448, R: 0.370, F1: 0.409

Citation

If you use this model, please cite like below:

@misc{hong2024dimbreminingscientificliterature,
      title={DiMB-RE: Mining the Scientific Literature for Diet-Microbiome Associations}, 
      author={Gibong Hong and Veronica Hindle and Nadine M. Veasley and Hannah D. Holscher and Halil Kilicoglu},
      year={2024},
      eprint={2409.19581},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2409.19581}, 
}