The Rise of Computational Anthropology

The Rise of Computational Anthropology
1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

Technology review via ACM Technews, reports that “Microsoft researchers in Beijing are using location-based social networks to study changes in human patterns of behavior in time and space. Using data from a Chinese location-based social network called, which is similar to Foursquare, the team downloaded more than 1.3 million location check-ins from five major cities in China. Because the data includes the users’ hometowns, the researchers could determine whether check-ins took place in a user’s own city or elsewhere.

The researchers used the training data to learn local and non-local mobility patterns and visit location popularity. They developed an algorithm that predicts a user’s next location based on current location and whether the person is local. The most effective results came from analyzing a person’s behavior patterns and estimating the extent to which this person resembles a local user, thus creating an indigenization coefficient that helps forecast future mobility patterns.

Urban Mobility

The researchers say their method significantly surpasses the performance of mixed algorithms that use only individual visiting history and location popularity. Beyond helping businesses target travelers or local people, the algorithm could be used to monitor changes in a person’s mobility patterns over time, which could help anthropologists study migration and how immigrants assimilate into a local community and contribute to the nascent science of computational anthropology.”

Article: Indigenization of Urban Mobility Zimo Yang, Nicholas Jing Yuan, Xing Xie, Defu Lian, Yong Rui, Tao Zhou, Cornell University. Submitted May 2014.

Photo Credits: 2008 Beijing Opening Ceremonies @ Ditan Park by kris krüg / FlickR