Finding Places in Text
slides.com/andrewjanco/finding-places
Google Pinpoint
Most geocoding software is built for present-day place names.
Geocoding
Need to focus by time period and region
Geocoding
Context & Domain
- Text as a sequence of characters
- Natural language processing
- Tokenization
- Pattern Matching
Search
from spacy.lang.de import German
nlp = German()
doc = nlp("Berlin ist eine Stadt in Deutschland.")
for token in doc:
print(token.i, token.text)
0 Berlin
1 ist
2 eine
3 Stadt
4 in
5 Deutschland
6 .
whgazetteer.org