Geo- social Network Analysis and Applica5ons Cecilia Mascolo joint work with C. Brown, A. Noulas and S. Scellato Big Data and Social Media Workshop February 2013, Glasgow, UK.
My Interests
Geo- social Network Analysis can lead to insights into social behaviour
Foursquare Places (London)
Foursquare Dynamics
What can we study with this type of data? Rela5onship of friendship and distance Rela,onship of interac,on and distance Human mobility and role of places Models for geo- social network evolu5on Rela,onship of communi,es and space Applica5ons: Recommenda5ons, adver5sement Urban planning Epidemics control
Effect of geography over social interac,on
How does geography affect interac5on?
Tuen5 Tuen5 dataset (Nov 11) 9.88 million registered users ~1 174 million friendship links 500 million messages in 3 months
Geographic Proper5es 100 % of contacts at distance greater than x km 90 % of friendships, interactions 80 70 60 50 40 30 20 10 60% of social connec5ons are at a distance of 10km. Only 10% of all distances between users are below 100km wall interactions friendships potential friendships 10 0 10 1 10 2 10 3 distance in km
Effect of geography on friendship Distance constrains social interac5on establishment but has a uniform effect on all interac5ons prob. of friendship 10 3 friendships δ 1.79 +ε; (ε=1.3e 06) wall directed δ 1.82 +ε; (ε=2.7e 07) 10 4 3 interactions δ 1.80 +ε; (ε=5e 08) 5 interactions 10 5 δ 1.81 +ε; (ε=2.7e 08) 10 interactions δ 1.83 +ε; (ε=1.1e 08) 10 6 10 7 10 8 10 0 10 1 10 2 10 3 distance δ in km
Effects of geography on interac5ons fraction of wall posts between friends Probability that a message is exchanged over a friendship link seems not to be very affected by distance prob. of interaction 10 1 10 2 wall directed (all interactions) 3 interactions 5 interactions 10 interactions 10 0 10 1 10 2 10 3 distance in km
Effect of Age Under 15 have 70% friends within 10km. They also have more friends and interact more.
Understanding human mobility
Samuel A. Stouffer Stouffer's law of intervening opportunities states, "The number of persons going a given distance is directly proportional to the number of opportunities at that distance and inversely proportional to the number of intervening opportunities." *! - Empirically proven using data for migrating families in the city of Cleveland." " - Is this true in our data?" * S. Stouffer (1940) Intervening opportunities: A theory relating mobility and distance, American Sociological Review 5, 845-867"
The importance of density Place density by far more important than city area size with respect to mean length of human movements (R 2 = 0.59 and 0.19 respec5vely).
Rank Distance u v
Rank universality three ci5es all ci5es The rank of all ci5es collapses to a single line. We have measured a power law exponent α = 0.84 ± 0.07
A new model for urban mobility
Simula5on Results...
Understanding communi,es and role of places
City- scale social networks We look at intra- city social networks People who have checked in at a place in a given city, and their friends who have also checked in at those places What do these place- based social networks look like?
City- scale social networks Degree: power- law distribu5on 10000 1000 Atlanta Boston Chicago Minneapolis Seattle Number of people 100 10 1 1 10 100 Number of friends
City- scale social networks Degree: power- law distribu5on Clustering coefficient: high (between 0.1 and 0.2, in random graph of the same size <0.001) Average shortest path length: small (about 4 hops), comparable to random graph (Clustering coeff. + average path length = small world ) Community structure (modularity > 0.4) Our city- level graphs have well- known structural proper,es of social networks.
The role of places Power- law distribu5on of place popularity like the degree distribu5on in the social network Number of places 10000 1000 100 Atlanta Boston Chicago Minneapolis Seattle 10 1 1 10 100 Number of visitors
Places vital for 5e forma5on Power- law distribu5on of place popularity Analyze triangles: >70% of triangles have one place shared between all three people à clustering around certain places These places could act as foci for,e forma,on
Role of categories Probability of friendship What is the role of place categories? 0.2 0.15 0.1 0.05 Atlanta Boston Chicago Minneapolis Seattle Probability of friendship between colocated people at places in each Foursquare category Some kinds of places are much more likely to reinforce friendship than others 0 Arts College Food Nightlife Outdoors Professional Residence Shop Transport Category
What can all this be used for? Friendship Recommenda,on: Exploi,ng Place Features in Link Predic,on on Loca,on- based Social Networks, Salvatore Scellato, Anastasios Noulas, Cecilia Mascolo. In Proceedings of 17th ACM Interna5onal Conference on Knowledge Discovery and Data Mining (KDD 2011). San Diego, USA. August 2011. The Importance of Being Placefriends: Discovering Loca,on- focused Online Communi,es. Chloë Brown, Vincenzo Nicosia, Salvatore Scellato, Anastasios Noulas, Cecilia Mascolo. In ACM SIGCOMM Workshop on Online Social Networks (WOSN 2012). Helsinki, Finland. August 2012. Place Recommenda,on: Mining User Mobility Features for Next Place Predic,on in Loca,on- based Services. Anastasios Noulas, Salvatore Scellato, Neal Lathia and Cecilia Mascolo. In Proceedings of IEEE Interna5onal Conference on Data Mining (ICDM 2012). Short Paper. Brussels, Belgium. December 2012. A Random Walk Around the City: New Venue Recommenda,on in Loca,on- Based Social Networks. Anastasios Noulas, Salvatore Scellato, Neal Lathia and Cecilia Mascolo. In Proceedings of ASE/IEEE Interna5onal Conference on Social Compu5ng (SocialCom). Amsterdam, The Netherlands. September 2012.
What can all this used for? More Modelling: Talking Places: Modelling and Analysing Linguis,c Content in Foursquare. Sandro Bauer, Anastasios Noulas, Diarmuid Ó Séaghdha, Stephen Clark and Cecilia Mascolo. In Proceedings of ASE/IEEE Interna5onal Conference on Social Compu5ng (SocialCom). Amsterdam, The Netherlands. September 2012. Exploi,ng Foursquare and Cellular Data to Infer User Ac,vity in Urban Environments. Anastasios Noulas, Cecilia Mascolo and Enrique Frias- Mar5nez. In Proceedings of 14th Interna5onal Conference on Mobile Data Management (MDM 2013). Milan, Italy. June 2013. Evolu,on of a Loca,on- based Online Social Network: Analysis and Models. Mili5adis Allamanis, Salvatore Scellato and Cecilia Mascolo. In Proceedings of ACM Internet Measurement Conference (IMC 2012). Boston, MA. November 2012.
References The length of bridge,es: structural and geographic proper,es of online social interac,ons. Yana Volkovich, Salvatore Scellato, David Laniado, Cecilia Mascolo and Andreas Kaltenbrunner. In Proceedings of Sixth Interna5onal AAAI Conference on Weblogs and Social Media (ICWSM 2012). Full Paper. Dublin, Ireland. June 2012. Far from the eyes, close on the Web: impact of geographic distance on online social interac,ons. Andreas Kaltenbrunner, Salvatore Scellato, Yana Volkovich, David Laniado, Dave Currie, Erik J. Jutemar, Cecilia Mascolo. In ACM SIGCOMM Workshop on Online Social Networks (WOSN 2012). Helsinki, Finland. August 2012. A tale of many ci,es: universal pawerns in human urban mobility. Anastasios Noulas, Salvatore Scellato, Renaud Lambiove, Massimiliano Pon5l, Cecilia Mascolo. In PLoS ONE. PLoS ONE 7(5): e37027. doi:10.1371/ journal.pone.0037027 Further work on place/communi5es, in prepara5on.
Thanks! Ques5ons? Cecilia Mascolo @cecim cecilia.mascolo@cl.cam.ac.uk www.cl.cam.ac.uk/users/cm542