Skill Extraction Datasets
Collection
Datasets for skill extraction from job ads, annotated with ESCO skills on sentence or span level. • 5 items • Updated • 2
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
The TECHWOLF subset, although smaller, represents a more generic distribution of job descriptions and skill spans. ESCO skills are directly annotated on the full sentence level, thus omitting the intermediate span identification step. ESCO v1.1.0 is used.
This dataset is part of a three-part evaluation dataset for skill extraction:
If you use this dataset, please include the following reference:
@article{decorte2023extreme,
title={Extreme multi-label skill extraction training using large language models},
author={Decorte, Jens-Joris and Verlinden, Severine and Van Hautte, Jeroen and Deleu, Johannes and Develder, Chris and Demeester, Thomas},
journal={arXiv preprint arXiv:2307.10778},
year={2023}
}