This post will be a bit of a quick hitter -
I scraped Twitter for tweets containing #[city] for major US cities and threw together a quick visualization in Tableau:
That is all.
Quick Notes on Methodology:
Throwing out punctuation, special characters, and emojis (although I might bring in emojis later if I can figure out how to work with them in python).
Throwing out “stop words” - commonly occurring words such as [“the”, “and”, “or”] that don’t add value.
Throwing out words that occur in less than 0.5% of tweets (these would show up as tiny in the word cloud anyway).
NOT throwing out numbers - I thought about it, but frequently reoccurring numbers actually seem pretty interesting to me.
Scraped roughly 3K tweets per #city per day - planning to increase this going forward.