We could add a function to automatically download text for a given keyword (or set of keywords) from one of a pre-selected set of sources (e.g. wikipedia, twitter, youtube comments, etc.). E.g.:
words = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='twitter', text_params={'n_tweets': 1000})
returns a list (length 4) of lists of tweets about each topic. Or
words = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='wikipedia')
would return a list (length 4) of strings, each containing the text from the corresponding wikipedia article.
Both of these should be able to be plotted or analyzed via hyp.plot(words).
As a shortcut, the user could also pass:
hyp.plot(['apple', 'banana', 'cat', 'dog'], text_source='twitter'), which would (with that one command):
Call words = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='twitter')
We could add a function to automatically download text for a given keyword (or set of keywords) from one of a pre-selected set of sources (e.g. wikipedia, twitter, youtube comments, etc.). E.g.:
words = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='twitter', text_params={'n_tweets': 1000})returns a list (length 4) of lists of tweets about each topic. Orwords = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='wikipedia')would return a list (length 4) of strings, each containing the text from the corresponding wikipedia article.Both of these should be able to be plotted or analyzed via
hyp.plot(words).As a shortcut, the user could also pass:
hyp.plot(['apple', 'banana', 'cat', 'dog'], text_source='twitter'), which would (with that one command):words = hyp.tools.textlookup(['apple', 'banana', 'cat', 'dog'], text_source='twitter')wordsExample twitter scraper: https://github.com/ContextLab/storytelling-with-data/blob/master/data-stories/twitter-finance/twitter-finance.ipynb