Quantifying Social and Opportunistic Behavior in Email Networks

Email graphs have been used to illustrate the general properties of social networks of communication and collaboration. However, increasingly, the majority of Internet traffic reflects opportunistic rather than symbiotic social relations. Here we use email data drawn from a large university to construct directed graphs of email exchange that quantify the differences between social and opportunistic behavior, represented by legitimate messages and spam, respectively. We show that while structural characteristics typical of other social networks are shared to a large extent by the legitimate component, they are not characteristic of opportunistic traffic. To complement the graph analysis, which suffers from incomplete knowledge of users external to the domain, we study temporal patterns of communication to show dynamical properties of email traffic. The results indicate that social email traffic has lower entropy (higher structural information) than opportunistic traffic for periods covering both working and non-working hours. We see in general that both social and opportunistic traffics are not random, and that social email shows stronger temporal structure with a high probability for long silences and bursts of a few messages. These findings offer insights into the fundamental differences between social and opportunistic behavior in email networks, and may generalize to the structure of opportunistic social relations in other environments.

[1]  A. U.S.,et al.  Predictability , Complexity , and Learning , 2002 .

[2]  E. Wenger,et al.  Communities and technologies , 2003 .

[3]  S. Bornholdt,et al.  Scale-free topology of e-mail networks. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[4]  Stephanie Forrest,et al.  Email networks and the spread of computer viruses. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  Dimitrina S. Dimitrova,et al.  Computer Networks as Social Networks: Collaborative Work, Telework, and Virtual Community , 1996 .

[6]  M E J Newman.  Assortative mixing in networks. , 2002, Physical review letters.

[7]  M. Newman,et al.  Why social networks are different from other types of networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[8]  N.B. Shroff,et al.  Joint resource allocation and base-station assignment for the downlink in CDMA networks , 2006, IEEE/ACM Transactions on Networking.

[9]  Albert-László Barabási,et al.  The origin of bursts and heavy tails in human dynamics , 2005, Nature.

[10]  Steps to Reducing Unwanted Traffic on the Internet Workshop, SRUTI'05, Cambridge, MA, USA, July 7, 2005 , 2005, SRUTI.

[11]  Virgílio A. F. Almeida,et al.  Improving Spam Detection Based on Structural Similarity , 2005, SRUTI.

[12]  A. Arenas,et al.  Community analysis in social networks , 2004 .

[13]  Sergei Maslov,et al.  Hierarchy measures in complex networks. , 2003, Physical review letters.

[14]  Diego Garlaschelli,et al.  Patterns of link reciprocity in directed networks. , 2004, Physical review letters.

[15]  Stuart I. Feldman.  ACM's past helps steer its future , 2007, CACM.

[16]  Greg Goth.  The Long Tail Takes Over Music , 2007, IEEE Internet Computing.

[17]  P. Oscar Boykin,et al.  Leveraging social networks to fight spam , 2005, Computer.

[18]  Jean-Pierre Eckmann,et al.  Entropy of dialogues creates coherent structures in e-mail traffic. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[19]  S H Strogatz,et al.  Random graph models of social networks , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[20]  William Bialek,et al.  Spikes: Exploring the Neural Code , 1996 .

[21]  A Díaz-Guilera,et al.  Self-similar community structure in a network of human interactions. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.