Skip to content
What Twitter reveals about the differences between cities and the monoculture of the Bay Area
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
images metrotwitter May 31, 2019
vocabs metrotwitter May 31, 2019
.gitignore metrotwitter May 31, 2019
README.md metrotwitter May 31, 2019
analysis.py metrotwitter May 31, 2019
config.py metrotwitter May 31, 2019
emojis.txt metrotwitter May 31, 2019
keywords.ipynb metrotwitter May 31, 2019
processing.py metrotwitter May 31, 2019
requirements.txt metrotwitter May 31, 2019
utils.py metrotwitter May 31, 2019
wordclouds.py metrotwitter May 31, 2019

README.md

MetroTwitter

Analysis of Twitter data in 13 English-speaking metropolitan areas (96K users and 180M tweets).

  • US Cities (9): Atlanta, Austin, Bay Area, Boston, Chicago, Washington DC, LA, NYC, Seattle
  • Australian Cities (2): Melbourne, Sydney
  • Canadian Cities (1): Toronto
  • UK Cities (1): London

Plus 223K users that aren't in these areas which are collectively put in 'Other'.

For more information, see the blog post:

MetroTwitter - What Twitter reveals about the differences between cities and the monoculture of the Bay Area

The blog post only contains a small, selected number of visualizations. For more visualization, download here.

I won't be distributing the data for this project to protect users' privacy. If you'd like to discuss the data, contact me through my website huyenchip.com.

How people in different cities describe themselves

In the metrotwitter_visualization folder, you can find word clouds that represent the most popular words in bios in each city. The indi folder visualizes each city independently. The duo folder visualizes the difference between two cities.

What people in different cities talk about

In the metrotwitter_visualization folder, you can find word clouds that represent the most popular words in tweets in each city. The indi folder visualizes each city independently. The duo folder visualizes the difference between two cities.

Ranking cities by popularity of keywords

In the keywords Jupyter notebook in this GitHub repo, there's the method rank_cities_by_keyword to visualize any keyword you want, either using bios or tweets.

You can also plots multiple keywords on the same plot using the method rank_cities_by_multiple_keywords.

Ranking keywords within a city

Also in the keywords Jupyter notebook, there's the method rank_keywords_in_city to rank the popularity of keywords within a city.

Unique popular emojis in each city

Most unique city

Compare two keywords

Just to get a sense of how popular one vs another.

You can’t perform that action at this time.