Chris Wong realized a video that shows in a map 3 days of geo-tagged tweets in New York. Using the streaming API of the service Chris Whong collected three complete days of geo-tagged tweets. The video does not only reconstruct the coordinates of individual tweets but also simulates the trajectory between tweet locations. The “static” tweets are depicted as blue points, and the “dynamic” ones are painted in yellow. Each frame of the video represents one actual minute of the life in New York. It is remarkable how the video does gives us an idea of the daily life using only geo-tagged points. For example the rush hours can be clearly identified without looking at the actual time.
However, you need to take this video with a grain of salt. To begin with, it is clear that the scope of this video is limited to the twitter users located in the metropolitan region of New York City. Then, you have to limit once again the scope to the amount of persons that activated the geo-tagging function of twitter. Finally, according to this (( http://www.uncertaintyofidentity.com/resources/Agile2013-Final-Paper.pdf )) paper the API of twitter just provide a sample including approximately 1% of the tweets generated at any given time point. This is why the number of tweets is remarkably low (79,914) for three days of tweets in New York.
Now, I’m not trying to discredit the video, I just want to point out some facts that you should take into consideration if you watch it. Having said that, the video is a very interesting piece of information, specially if you think that it provides at least an idea of the actual journeys that the inhabitants undertake at any given time point. This information can be extremely valuable for the transport administration, as it could help to identify patterns of mobility that guide future improvements of the service.
Chris Wong is a Research Assistant at the Rudin center for transportation policy and management of the New York University.