A Community-Aware Approach to Minimizing Dissemination in Graphs

Given a graph, can we minimize the spread of an entity (such as a meme or a virus) while maintaining the graph’s community structure (defined as groups of nodes with denser intra-connectivity than inter-connectivity)? At first glance, these two objectives seem at odds with each other. To minimize dissemination, nodes or links are often deleted to reduce the graph’s connectivity. These deletions can (and often do) destroy the graph’s community structure, which is an important construct in real-world settings (e.g., communities promote trust among their members). We utilize rewiring of links to achieve both objectives. Examples of rewiring in real life are prevalent, such as purchasing products from a new farm since the local farm has signs of mad cow disease; getting information from a new source after a disaster since your usual source is no longer available, etc. Our community-aware approach, called constrCRlink (short for Constraint Community Relink), preserves (on average) \(98.6\%\) of the efficacy of the best community-agnostic link-deletion approach (namely, NetMelt \(^{+}\)), but changes the original community structure of the graph by only \(4.5\%\). In contrast, NetMelt \(^{+}\) changes \(13.6\%\) of the original community structure.

[1]  Christos Faloutsos,et al.  Epidemic thresholds in real networks , 2008, TSEC.

[2]  Hanghang Tong,et al.  MET: A Fast Algorithm for Minimizing Propagation in Large Graphs with Small Eigen-Gaps , 2015, SDM.

[3]  M. Newman,et al.  Robustness of community structure in networks. , 2007, Physical review. E, Statistical, nonlinear, and soft matter physics.

[4]  Steven B. Andrews,et al.  Structural Holes: The Social Structure of Competition , 1995, The SAGE Encyclopedia of Research Design.

[5]  Jure Leskovec,et al.  Empirical comparison of algorithms for network community detection , 2010, WWW '10.

[6]  C. Lanczos.  An iteration method for the solution of the eigenvalue problem of linear differential and integral operators , 1950 .

[7]  Alessandro Flammini,et al.  Optimal network clustering for information diffusion , 2014, Physical review letters.

[8]  Hanghang Tong,et al.  Make It or Break It: Manipulating Robustness in Large Networks , 2014, SDM.

[9]  Samarth Swarup,et al.  Blocking Simple and Complex Contagion by Edge Removal , 2013, 2013 IEEE 13th International Conference on Data Mining.

[10]  Jean-Loup Guillaume,et al.  Fast unfolding of communities in large networks , 2008, 0803.0476.

[11]  Santo Fortunato,et al.  Community detection in graphs , 2009, ArXiv.

[12]  Alessandro Flammini,et al.  Erratum: Optimal Network Modularity for Information Diffusion [Phys. Rev. Lett. 113, 088701 (2014)] , 2014 .

[13]  M E J Newman,et al.  Community structure in social and biological networks , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[14]  Michalis Faloutsos,et al.  Threshold Conditions for Arbitrary Cascade Models on Arbitrary Networks , 2011, IEEE ICDM 2011.

[15]  Sudip Saha,et al.  Approximation Algorithms for Reducing the Spectral Radius to Control Epidemic Spread , 2015, SDM.

[16]  Ramanathan V. Guha,et al.  Information diffusion through blogspace , 2004, SKDD.

[17]  Michalis Faloutsos,et al.  Eigen-Optimization on Large Graphs by Edge Manipulation , 2016, ACM Trans. Knowl. Discov. Data.

[18]  Ravi Kumar,et al.  On the Bursty Evolution of Blogspace , 2003, WWW '03.

[19]  Yao Zhang,et al.  Near-Optimal Algorithms for Controlling Propagation at Group Scale on Networks , 2016, IEEE Transactions on Knowledge and Data Engineering.

[20]  Martin Rosvall,et al.  Maps of random walks on complex networks reveal community structure , 2007, Proceedings of the National Academy of Sciences.

[21]  Michalis Faloutsos,et al.  Gelling, and melting, large graphs by edge manipulation , 2012, CIKM.

[22]  Alessandro Vespignani,et al.  Epidemic spreading in scale-free networks. , 2000, Physical review letters.