Solving the contamination minimization problem on networks for the linear threshold model

Abstract

We address the problem of minimizing the spread of undesirable things, such as computer viruses and malicious rumors, by blocking a limited number of links in a network. This optimization problem called the contamination minimization problem is, not only yet another approach to the problem of preventing the spread of contamination by removing nodes in a network, but also a problem that is converse to the influence maximization problem of finding the most influential nodes in a social network for information diffusion. We adapted the method which we developed for the independent cascade model, known for a model for the spread of epidemic disease, to the contamination minimization problem under the linear threshold model, a model known for the propagation of innovation which is considerably different in nature. Using large real networks, we demonstrate experimentally that the proposed method significantly outperforms conventional link-removal methods. © 2008 Springer Berlin Heidelberg.