Dear all,
I have a database with more than one million tweets (about WW1). What I'd like to do now is to recognize dates and traces of time.
Has anyone an idea of how to do this? (In French and English).
Best
Which Named Entity Recognition Tool?
(4 posts) (3 voices)-
Posted 1 year ago Permalink
-
Replying to @inactinique's post:
Bump, because although I don't know the/an answer, I'm interested.
Posted 1 year ago Permalink -
I checked some of the bigger APIs (OpenCalais, AlchemyAPI) and they don't seem overly interested in flexible parsing of dates. If a bit of coding is possible, you might have more luck with GATE or StanfordNLP, and in particular the SUTime module that can be tested here: http://nlp.stanford.edu:8080/sutime/process (the bad news is that I suspect this language model is limited to English). There's probably an Python NLTK module as well, but nothing I could find after a quick search was particularly well-suited for detecting "traces of time". Please do let us know if you find something! I could maybe be bribed to look at adding this to Voyant ;)
Posted 1 year ago Permalink -
Thanks Stéfan. Curiously, I haven't seen your answer before. There is a French app, called Tropes, unfortunately not open source, that can be used (an English version exists). But it is very limited. I'll look at StanfordNLP or NLTK if I can do something with it.
Posted 10 months ago Permalink
Reply
You must log in to post.