Dmjdxb/batterybert-electrocatalyst-ner-v3

BatteryBERT Electrocatalyst NER v3

Fine-tuned NER model for extracting entities from electrocatalyst research literature.

Model Description

This model identifies 6 entity types in scientific text about electrocatalysts:

Entity	Description	Example
MATERIAL	Catalyst materials, compounds	IrO₂, NiFe-LDH, Pt/C
CONDITION	Experimental conditions	1.6 V vs RHE, 80°C, pH 7
METRIC	Performance measurements	10 mA/cm², 45 mV/dec
PROCESS	Synthesis/characterization methods	electrodeposition, annealing
ELECTROLYTE	Electrolyte solutions	0.5 M H₂SO₄, 1 M KOH
DURATION	Time periods	100 h, 1000 cycles

Performance

Entity	Precision	Recall	F1-Score
CONDITION	0.66	0.75	0.70
DURATION	0.77	0.89	0.83
MATERIAL	0.62	0.81	0.70
METRIC	0.58	0.79	0.67
PROCESS	0.76	0.69	0.72
ELECTROLYTE	0.28	0.21	0.24
Overall	0.64	0.73	0.68

Usage

from transformers import pipeline

ner = pipeline("ner", model="Dmjdxb/batterybert-electrocatalyst-ner-v3", aggregation_strategy="first")

text = "IrO₂ showed 10 mA/cm² at 1.6 V vs RHE in 0.5 M H₂SO₄ after 100 h."
entities = ner(text)

for e in entities:
    print(f"{e['entity_group']}: {e['word']} ({e['score']:.0%})")

Training Data

• Sentences: 1,824
• Entities: 2,681
• Papers: 140 electrocatalysis research articles

Training Details

• Base model: batterydata/batterybert-uncased
• Epochs: 3
• Learning rate: 2e-5
• Batch size: 16

Intended Use

Information extraction from scientific literature on OER/HER electrocatalysts and water splitting.

Citation

@misc{batterybert-electrocatalyst-ner-v3,
  author = {David Johnson},
  title = {BatteryBERT Electrocatalyst NER v3},
  year = {2026},
  publisher = {HuggingFace},
  url = {https://huggingface.co/Dmjdxb/batterybert-electrocatalyst-ner-v3}
}