I need to answer those two questions:

  1. Are there any locations (in the sense of place: "capital of France", "South of Spain", "London", "New Zealand", etc. ) mentioned in a particular text?
  2. If so: which locations are mentioned?

And I'm not quite sure what's the best tooling / process for it. And maybe what keywords I have to watch out for.

I heard good things about word2vec but I'm not quite sure whether it is suitable here.

  • Hey, welcome to SE! I may be out of depth here but what do you exactly mean by locations? May 18 '19 at 19:02
  • Location in the sense of a place. For example "capital of France", "South of Spain", "London", "New Zealand", etc. May 18 '19 at 21:08

For your first question, identifying locations in a piece of text isn't too difficult. The process of Named Entity Recognition (NER) does it. Depending on the text, you can use libraries like SpaCy and Apache NLP (better for simpler tasks) OR you can train your own models (better suited for more domain-specific tasks like identifying names of football stadiums).

However, in the text like 'capital of France' libraries like SpaCy would identify 'France' as a location and not the entire piece 'capital of France'.

  • 1
    Awesome. Thank you! Sounds perfect! May 22 '19 at 13:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.