Lost in the City: Revisiting Milgram's Experiment in the Age of Social Networks

As more and more users access social network services from smart devices with GPS receivers, the available amount of geo-tagged information makes repeating classical experiments possible on global scales and with unprecedented precision. Inspired by the original experiments of Milgram, we simulated message routing within a representative sub-graph of the network of Twitter users with about 6 million geo-located nodes and 122 million edges. We picked pairs of users from two distant metropolitan areas and tried to find a route between them using local geographic information only; our method was to forward messages to a friend living closest to the target. We found that the examined network is navigable on large scales, but navigability breaks down at the city scale and the network becomes unnavigable on intra-city distances. This means that messages usually arrived to the close proximity of the target in only 3–6 steps, but only in about 20% of the cases was it possible to find a route all the way to the recipient, in spite of the network being connected. This phenomenon is supported by the distribution of link lengths; on larger scales the distribution behaves approximately as , which was found earlier by Kleinberg to allow efficient navigation, while on smaller scales, a fractal structure becomes apparent. The intra-city correlation dimension of the network was found to be , less than the dimension of the distribution of the population.

[1]  Christopher M. Danforth,et al.  The Geography of Happiness: Connecting Twitter Sentiment and Expression, Demographics, and Objective Characteristics of Place , 2013, PloS one.

[2]  Jasmine Novak,et al.  Geographic routing in social networks , 2005, Proc. Natl. Acad. Sci. USA.

[3]  D. Watts,et al.  An Experimental Study of Search in Global Social Networks , 2003, Science.

[4]  Hawoong Jeong,et al.  Modeling the Internet's large-scale topology , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[5]  Yutaka Matsuo,et al.  Earthquake shakes Twitter users: real-time event detection by social sensors , 2010, WWW '10.

[6]  István Csabai,et al.  A multi-terabyte relational database for geo-tagged social network data , 2013, 2013 IEEE 4th International Conference on Cognitive Infocommunications (CogInfoCom).

[7]  D. ben-Avraham,et al.  Kleinberg navigation in fractal small-world networks. , 2006, Physical review. E, Statistical, nonlinear, and soft matter physics.

[8]  Lars Backstrom,et al.  The Anatomy of the Facebook Social Graph , 2011, ArXiv.

[9]  Guoqiang Shen,et al.  Fractal dimension and fractal growth of urbanized areas , 2002, Int. J. Geogr. Inf. Sci..

[10]  Zengru Di,et al.  Possible origin of efficient navigation in small worlds. , 2011, Physical review letters.

[11]  Jacob Goldenberg,et al.  Distance Is Not Dead: Social Interaction and Geographical Distance in the Internet Era , 2009, ArXiv.

[12]  Etienne Huens,et al.  Geographical dispersal of mobile communication networks , 2008, 0802.2178.

[13]  Christopher M. Danforth,et al.  Twitter reciprocal reply networks exhibit assortativity with respect to happiness , 2011, J. Comput. Sci..

[14]  Sándor Laki,et al.  On the spatial properties of internet routes , 2012, Comput. Networks.

[15]  M E J Newman,et al.  Identity and Search in Social Networks , 2002, Science.

[16]  Brandon Van Der Heide,et al.  A social network as information: The effect of system generated reports of connectedness on credibility on Twitter , 2012, Comput. Hum. Behav..

[17]  P. Gloor,et al.  Predicting Stock Market Indicators Through Twitter “I hope it is not as bad as I fear” , 2011 .

[18]  Jon M. Kleinberg,et al.  The small-world phenomenon: an algorithmic perspective , 2000, STOC '00.

[19]  F. Chung,et al.  The average distances in random graphs with given expected degrees , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[20]  Cécile Caretta Cartozo,et al.  Extended navigability of small world networks: exact results and new insights. , 2009, Physical review letters.

[21]  Péter Mátray,et al.  Measuring the dimension of partially embedded networks , 2013, ArXiv.

[22]  S. Appleby.  Multifractal Characterization of the Distribution Pattern of the Human Population , 2010 .

[23]  S. Havlin,et al.  Dimension of spatially embedded networks , 2011 .

[24]  Stanley Milgram,et al.  An Experimental Study of the Small World Problem , 1969 .

[25]  Alan M. Frieze,et al.  Random graphs , 2006, SODA '06.

[26]  Jure Leskovec,et al.  Friendship and mobility: user movement in location-based social networks , 2011, KDD.

[27]  Marián Boguñá,et al.  Navigability of Complex Networks , 2007, ArXiv.

[28]  Zohreh Azimifar,et al.  Degrees of Separation in Social Networks , 2011, SOCS.

[29]  Sharon L. Milgram,et al.  The Small World Problem , 1967 .

[30]  Christopher M. Danforth,et al.  Happiness and the Patterns of Life: A Study of Geolocated Tweets , 2013, Scientific Reports.

[31]  P. Grassberger,et al.  Measuring the Strangeness of Strange Attractors , 1983 .

[32]  Christopher M. Danforth,et al.  Temporal Patterns of Happiness and Information in a Global Social Network: Hedonometrics and Twitter , 2011, PloS one.