5. Convert LCSH headings to GeoNames

Set-up

To complete this step, you need:

• Access to the command line
• Python 3+ with:
  • Beautiful Soup 4
  • numpy
  • pandas
  • requests
  • rdflib
• Scripts:
  • addGeoNames.py
  • convertGeoNamesFromLCNAF.py

---

Run addGeoNames.py

Transformations in this script:

• In step 3, the script extractMARCToGeoCSV.py extracted geographic-related components of subject headings from the 600 fields of the MARC record. It divided these subjects into likely FAST subject headings ("Maryland--Baltimore") and likely LCNAF subject headings ("Baltimore (Md.)"). This script transforms those geographic headings into hierarchical string headings based on GeoNames. Therefore, "Maryland--Baltimore" and "Baltimore (Md.)" are both transformed to Baltimore, Maryland, United States.
  • This is done by finding "sameAs" references to GeoNames within the FAST authority files.