UPDATE MARCH 2019: This repo is depreciated due to API changes by Foursquare to stop scraping of this particular metadata. This repository will remain up for posterity.

A Foursquare data scraper that gathers all venues within a specified geographic area. The data retrieved can then be used for statistical analysis and fun data visualizations.

The script queries the official Foursquare API (with a user-provided client_id and client_secret from a Foursquare application) to search for venues, and then query additional metadata for each unique venue.

The specified area is defined by a geographic bounding box (by latitude and longitude). Since the API can only return up to 50 venues per search, the bounding box is broken into sub-bounding boxes as a form of grid search (by default, the grid_size is 100, so 100x100=10000 searches total).

The output file is a CSV with the following fields:

• id — The Foursquare ID for the venue.
• name — The name of the venue.
• categories — The list of categories of the venue.
• lat — Latitude
• long — Longitude
• num_checkins — Number of Foursquare Checkins
• num_likes — Number of Foursquare Likes
• price — Price Tier between 1-4 (i.e. $-$$$$)
• rating — Rating for the venue.
• num_ratings — Number of ratings for the venue.
• url_venue — URL of the venue.
• url_foursquare — URL for the Foursquare venue.

A demo CSV of the Top 100 San Francisco venues (by number of checkins) is available in foursquare_sf_sample.csv.

Set the parameters in config.yaml as appropriate, then run foursquare_venues.py.

The Foursquare API has a daily rate limit of 1,000 requests/day for free accounts, but with verification, the rate limit upgrades to 100,000/day. You can test the script with a small bounding box/low grid size to see if it fits your needs before upgrading.

Both of the API endpoints used in this script have a rate limit of 5,000 requests/hour. Once the script hits that limit, it'll sleep for an hour. Plan scraping accordingly.

This script follows the specifications of the Foursquare API, and does not circumvent the API and its rate limits. If you're using the data for projects other than data analysis, make sure to follow the Foursquare Terms of Use.

Max Woolf (@minimaxir)

Max's open-source projects are supported by his Patreon. If you found this project helpful, any monetary contributions to the Patreon are appreciated and will be put to good creative use.

MIT