Automatically identify and categorize place names

[JonathanReeve]

1. Do some NER to find place names (SpaCy? Entity type: place) Here is some documentation from SpaCy
2. Automatically look up that place name in DBPedia, using the DBPedia API and the Python requests library.

• Here's an example JSON object returned.
• And here's the page

3. Use that data to include latitude, longitude, description, etc. for each (real) place.
4. Make a heuristic for categorizing place names.

[JonathanReeve]

@matthewkumar, want to give this a shot?

[JonathanReeve]

This is very closely related to #27, and is partially complete in #41.

[JonathanReeve]

See also open-editions/corpus-joyce-portrait-TEI#90. It's best to have place name metadata in a separate file, say, places.xml, and each place name in the text is referenced by an xml id. Here's some TEI documentation

[JonathanReeve]

We should also integrate @muziejus's Wandering Rocks data: https://github.com/muziejus/wandering-rocks/blob/master/data/instances.csv

[workshub]

A user started working on this issue via WorksHub.

[workshub]

@Freitas-Mp started working on this issue via WorksHub.

[JonathanReeve]

Hi @Freitas-Mp! Could you tell me what you had in mind for this issue? I may be able to help you think through it.

[workshub]

@baoduong started working on this issue via WorksHub.

[workshub]

A user started working on this issue via WorksHub.

[workshub]

@bragaji started working on this issue via WorksHub.