⚠️ Note: This model has been superseded by newer versions fine-tuned on top of ModernBERT architectures.
Check out the improved models here:

• 🔬 ModernBERT-base-biomedical-ner
• 🧬 ModernBERT-large-biomedical-ner

These newer models offer enhanced performance for biomedical NER tasks and are recommended for most use cases.

Model Card: bert-base-cased-biological-ner

Model Details

• Model Name: bert-base-cased-biomedical-ner
• Model Architecture: BERT (Bidirectional Encoder Representations from Transformers)
• Pre-trained Model: bert-base-cased
• Fine-tuned on: SourceData Dataset

Model Description

The bert-base-cased-biomedical-ner is a fine-tuned variant of the BERT (Bidirectional Encoder Representations from Transformers) model, designed specifically for the task of Named Entity Recognition (NER) in the biomedical domain. The model has been fine-tuned on the SourceData Dataset, which is a substantial and comprehensive biomedical corpus for machine learning and AI in the publishing context.

Named Entity Recognition is a crucial task in natural language processing, particularly in the biomedical field, where identifying and classifying entities like genes, proteins, diseases, and more is essential for various applications, including information retrieval, knowledge extraction, and data mining.

Intended Use

The bert-base-cased-biological-ner model is intended for NER tasks within the biomedical domain. It can be used for a range of applications, including but not limited to:

• Identifying and extracting biomedical entities (e.g., genes, proteins, diseases) from unstructured text.
• Enhancing information retrieval systems for scientific literature.
• Supporting knowledge extraction and data mining from biomedical literature.
• Facilitating the creation of structured biomedical databases.

Labels

Usage

from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer
import pandas as pd

tokenizer = AutoTokenizer.from_pretrained("Kushtrim/bert-base-cased-biomedical-ner")
model = AutoModelForTokenClassification.from_pretrained("Kushtrim/bert-base-cased-biomedical-ner")

ner = pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy='first')

text = "Add your text here"

results = ner(text)

pd.DataFrame.from_records(results)

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1

Development and attribution

• Original BERT architecture and base weights: Google BERT Team, through google-bert/bert-base-cased
• Biomedical NER fine-tuning, evaluation, packaging, and release: Kushtrim Visoka
• Model card and ongoing maintenance: Kushtrim Visoka

Acknowledgements

This work builds on google-bert/bert-base-cased and the work of its original authors. The upstream architecture and base weights remain the work of their respective creators. This repository documents the task-specific adaptation and release work carried out by Kushtrim Visoka and does not claim authorship of the original model. Dataset creators and tool authors referenced elsewhere in this card should also be cited according to their licenses and documentation.

Citation

If you use this fine-tuned model, please cite this repository and the original upstream model.

@misc{visoka_bert_base_cased_biomedical_ner_2026,
  author       = {Visoka, Kushtrim},
  title        = {BERT Base Cased Biomedical NER},
  year         = {2026},
  publisher    = {Hugging Face},
  url          = {https://huggingface.co/Kushtrim/bert-base-cased-biomedical-ner},
  note         = {Fine-tuned derivative of google-bert/bert-base-cased}
}