Multiple Topics Misinformation blocking in Online Social Networks

Abstract

Misinformation prevention has received much attention due to its important role to user community. However, recent studies ignore the influence of the topics of misinformation in the process of information dissemination. In fact, the spread of propagation of misinformation depends on their topics. Therefore, in order to improve the effectiveness of preventing false information, we need to consider the effect of topics in information dissemination.In this paper, we study the problem of misinformation blocking which considers topics of misinformation sources by removing a set of nodes, called MTMB problem. We show that MTMB is NP-hard and the objective function is a monotone and submodular function. Based on that, we propose a Greedy Algorithm (GA), which provides a approximation ratio of $\left( {1 - 1/\sqrt e } \right)$. We further propose a Scalable Greedy Algorithm (SGA), an efficient algorithm based on speeding up the GA by effective estimating the objective function. Experiments are conducted on networks showing the effectiveness and running time of the proposed algorithms which outperform other methods.