CZ Software Mentions: A large dataset of software mentions in the biomedical literature - Expanded 2024

We release a dataset of software mentions in open access biomedical papers published in bioRxv, medRxiv, or stored in euroPMC. The mentions are extracted with a trained BERT model. The dataset provides sources, context, metadata, software, and links.

This is a continuation of our previous dataset, https://datadryad.org/stash/dataset/doi:10.5061/dryad.6wwpzgn2c, based on an expanded set of papers

Methods
We used the following corpus of papers:

Source
Last updated
Number of unique DOIs

bioRxiv
 2023/08/23 
198247

medRxiv
 2023/08/24
43404

europePMC
2023/06/12
5289607

We run a BERT-based NER model on our corpus to extract plain-text software mentions. The model has been trained on the SoftCite dataset, uses the SciBERT model architecture and has an F1 score of 0.922. More details about this model can be found at https://github.com/chanzuckerberg/software-mention-extraction. We performed linking and disambiguation based on the procedures discussed in our paper https://arxiv.org/abs/2209.00693