TEI-NER Gazetteer: Historical Person and Place Name Database for Entity Linking
收藏资源简介:
External authority data databases for the TEI-NER-Pipeline — automatic entity linking for historical TEI editions. persons.db (~6.5M entries): Persons with GND/Wikidata IDs, name variants, life dates. Compiled from Wikidata SPARQL and GND/LOBID. Places.db (~524K entries): Places with Wikidata IDs, historical spellings, coordinates. Compiled from Wikidata, Wikipedia (historical name extraction), and GenWiki charter evidence. These databases serve as preload cache and lookup sources for the TEI-NER-Pipeline, reducing API calls by ~70% and enabling offline operation. The pipeline automatically links persons, places, and organizations in TEI-XML registers with authority data from GND and Wikidata, achieving 98.9% accuracy. Data sources: Wikidata (CC0), GND/LOBID (CC0), Wikipedia (CC BY-SA 4.0), GenWiki (CC BY-SA).



