Dataset · amirveyseh
Scientific Text Acronym Identification Corpus
amirveyseh/acronym_identification
Annotated English sentences for training models to find acronyms in documents.
What's inside
The dataset contains token-level annotations in English sentences indicating which words are acronyms. It includes training, validation, and test splits (10K to 100K examples total). Data is prepared for Shared Task 1 at the AAAI-21 workshop on scientific document analysis.
What you can do with it
- Training NER models (token-classification) for abbreviation extraction
- Preprocessing and parsing English scientific literature
- Tasks
- Token classification
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 20.8K
- Likes
- 27
How to load it
from datasets import load_dataset
ds = load_dataset("amirveyseh/acronym_identification")