I’ve always been a huge fan of mapping neighborhoods and perceived neighborhood boundaries. These often subjective, invisible boundaries are very useful in many applications and in my mind a huge, untapped opportunity (ping me if you want to go into business!). This article from Fast Company looks at mapping and analyzing boundaries created from foursquare data called Livehoods (http://livehoods.org/) – the research effort by some keen Carnegie Mellon U mobile lab students (http://mcom.cs.cmu.edu/) analyzed 18 million Foursquare check-ins to spot algorithmic relationships between the spots people frequent. According to the work… “Livehoods looks at the geographic distance between venues, but also a form of `social distance’ that measures the degree of overlap in the people that check-in to them”.
So what exactly is Livehoods? From the team… Using data such as tweets and check-ins, we are able to discover the hidden structures of the city with machine learning. Our techniques reveal a snap-shot of the dynamic areas the comprise the city, which we call Livehoods.
So, how are the neighborhood map boundaries defined? Quite simple in premise really. The team gathers data from foursquare social checkins. The shapes of Livehoods are determined by the patterns of people that check-in to them. If many of the same people check-in to two nearby locations, then these locations will likely be part of the same Livehood. Sounds good to me!! FYI, you can follow livehoods on Twitter @livehoods or via facebook https://www.facebook.com/livehoods. FYI, currently the team is hosting a facebook poll to determine which city will be mapped by livehood next.. Montreal seems to be doing a good job getting the votes!
Livehoods was recently featured in an interesting FastCompany article
See More at http://livehoods.org/
