- Download Standford NER from https://nlp.stanford.edu/software/CRF-NER.html#Download
- Extract to project base directory
Geopandas was difficult to install on my M1 macOS, this guide worked:
Contains arbitrarily truncated versions of the full texts, to be used for testing the NER script more quickly
This script performs NER on the texts in either full_texts
or short_texts
, it creates one CSV file per text, containing all identified Entities. Additionally, it searches each placename using the GHAP API and creates a map of the results.
This script searches names from test_csv.csv
using the GHAP API and attempts to disambiguate to a single location. It creates a single CSV of all outputs