This dataset was produced for the purpose of fine-tuning a Named Entity Recognition (NER) model with domain-specific knowledge relevant to the historic textile industry of the United Kingdom around the turn of the twentieth century.
The dataset has been formatted specifically for fine-tuning models from the GLiNER family.
The dataset was produced as part of the Science Museum Group's Congruence Engine research project.
