Fellowship Week 2: Visualizing Neighborhood Photos

This past week I went back to work after a long weekend hiking in Yosemite. With a clear head, I starting doing some reading and writing about key questions for my project in the Mission. I’ll be writing about some of these ideas and questions later this week. In the meantime, I wanted to explore social data in another San Francisco neighborhood, Civic Center.

My project will allow users to engage with a real-time map of recent human activity, including social media events like Foursquare checkins and Instagram photos. One tough question is whether data points related to neighborhood social events will be relevant to the users who briefly encounter the project. Ideally, there will be a shape to the data that with notable trends. After observing a trend or interesting data point, the user can then start to spot points of interests and even discover stories about the neighborhood. To test this question, I decide to do some experimentation.

I was thinking of neighborhood events and realized that I could look at the SF Pride Parade on Market St and Civic Center. I used the real-time Instagram API to capture photos taken during 12 hours of Pride festivities. Here’s a visualization of the resulting data:

CartoDB visualization of all Instagram photos taken during the 2015 Pride Parade.

You can play around with the data and visualization hereCartoDB instance of Pride Photos

There’s clearly a shape to the data, as the you can see the stream of photos grow from Financial District to Civic Center as the parade begins between 10:30 and 11am.

Photos at 11am during the SF Pride Parade

A couple of hours later, you can see other parts of Market heating up. People also start dispersing towards Civic Center and other blocks of the Tenderloin.


Here then, is the ongoing party just after the parade ended:


There were more that 10,000 photos taken during a 12-hour timeframe. I’ll be doing some additional spatial and statistical analysis on this data, such as how many photos are taken in the vicinity of other popular photos. One simple way to look at the data was to sort all the photos by comments and likes. Looking at hotspots on the map and popular photos, I was able to find photos that represent broader stories and trends at the event.

Most Popular Photos

Here the photo with the most (by far) likes (13,853) and comments (72):

And some other popular photos that I enjoyed.

View this post on Instagram

#SFPride from above.

A post shared by Chris Messina (@chris) on


2 thoughts on “Fellowship Week 2: Visualizing Neighborhood Photos

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s