advertisement
advertisement
advertisement

Scientists are using your Instagrams to understand cities

Believe it or not, your group selfies are going to be a gold mine for historians one day. Indeed, the value of detail-rich, location-tagged data from social networks like Foursquare and Twitter is already being mined for insights by scientists today. Their next frontier? Instagram. 

In a recent paper, Dutch researchers John D. Boy and Justus Uitermark outline how scientists can use data from Instagram’s API to analyze human behavior within cities and get a clearer picture of how communities and subcultures form and evolve in urban environments and the role that specific public places play in the process

While most of us think of Instagram as a social network or photo-sharing app, social scientists see it as “a participatory sensing system” because of the social and geospatial data generated by its 500 million users everyday.  By ingesting that data and breaking it down by social activity (likes, comments, follows), location and other details, researchers can start to get a new level of insight into the relationship we all share with the places we inhabit. 

“Which places facilitate encounters between members of different groups, and which are exclusive to members of the same group?” the researchers ask. They ranked places from most parochial to most cosmopolitan by employing a diversity measure known as the divergence index. In this chart of places in Amsterdam, the horizontal bar graph on the right shows the value of the divergence measure, with higher values indicating lower diversity.  The heatmap in the middle indicates the relative presence of different clusters of users.

Visualization of place diversity in Amsterdam. Via Boy, et al.

Previously, research into subcultures, communities and human activity within a city required much more onerous and tedious methods of research. And while the process of mining Instagram for urban insights is only just beginning (It’s easy to imagine how subfields of artificial intelligence like machine vision could help down the line), the potential future impact of this research is hard to miss. 

As more people flock to cities and the urban experience changes more rapidly, the slow-moving, traditional machinery of science will likely struggle to keep up. So it’s a good thing we’re creating a trove of useful data (in addition, of course, to all the data collected by advertisers, governments, and hackers) every time we lean in for a group shot or crouch down to frame that new mural in our phone’s viewfinder.  So go ahead, snap that selfie. If your friends give you shit, tell them it’s for science. JPT