Spine



 
 
  Welcome!
  SPINE (SParsification of Influence NEtworks) is a new algorithm for discovering the "backbone" of an influence network. Given a social network and a log of past propagations, we build an instance of the independent-cascade model that describes the propagations. We aim at reducing the complexity of that model, by preserving the network edges that are most likely to lead to propagations.
  
   Publications
  M. Mathioudakis, F. Bonchi, C. Castillo, A. Gionis. and A. Ukkonen. Sparsification of Influence Networks. To appear at KDD 2011 [pdf]
 
   Downloads
  
The project libraries, as well as sample datasets and instructions on how to use them, are available here.
   Team
  Francesco Bonchi 
Carlos Castillo 
Aris Gionis 
Michael Mathioudakis 
Antti Ukkonen 

Email: spine@cs.toronto.edu
  
Last Updated: August 2011