Quantifying locality in complex social networks Gábor Vattay Departmant of Physics of Complex Systems Eötvös University Budapest
Quantifying locality in complex social networks
Gábor Vattay
Departmant of Physics of Complex Systems
Eötvös University Budapest
My coworkers
• István Csabai István professzor
• Dániel Kondor Ph.D., MIT
• Eszter Bokányi graduate student
• László Dobos assistant professor
• Szüle János graduate student
• Kallus Zsófia graduate student
• Sebők Tamás graduate student
• Barankai Norbert graduate student
a SZOCIOLÓGIA mint
TERMÉSZETTUDOMÁNY
Twitter API
DB
PlanetLab nodes
User status updates
Indexed geo data
User connections graph
User Graph Discovery
Tool
Data Processing Framework
Our database2012-2014
4.0 Billion tweets
1.6 Billion GPS
130 Million users
Twitter friendship
Top 6 Million GPS
122 Million Friendships
Twitter friendship map @elte
Milgram’s small world experiment
Figure 4. Number of steps needed to reach the proximity of target users.
Szüle J, Kondor D, Dobos L, Csabai I, et al. (2014) Lost in the City: Revisiting Milgram's Experiment in the Age of Social Networks. PLoS ONE 9(11): e111973. doi:10.1371/journal.pone.0111973http://www.plosone.org/article/info:doi/10.1371/journal.pone.0111973
Clustering algorithms
Table 1. Regional graphs.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 2. Clustering of the United Kingdom.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 3. Clustering of the subgraph of Canada.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 4. Clustering of the subgraph of the Continental US.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 5. Second level partitioning of the Western US cluster.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 6. Clustering of the countries of South America.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 7. Clustering of the 28 member countries of the European Union combined with second-level results.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 8. Clustering of the countries of the European continent combined with second-level results.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 9. Communities formed in Switzerland.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 10. Communities formed in Cyprus and Greece.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 11. Communities formed in Germany and Turkey.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 12. Clustering of the Former Yugoslavia.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 13. Clustering of Spain.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 14. Clustering of France.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 15. Clustering of Germany.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 16. Clustering of the region of Southeast Asia with inclusion of China and Japan.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 17. Clustering of Japan.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
Fig 18. Clustering of India.
Kallus Z, Barankai N, Szüle J, Vattay G (2015) Spatial Fingerprints of Community Structure in Human Interaction Network for an Extensive Set of Large-Scale Regions. PLoS ONE 10(5): e0126713. doi:10.1371/journal.pone.0126713http://journals.plos.org/plosone/article?id=info:doi/10.1371/journal.pone.0126713
THANK YOU!