Scalable Rumor Source Detection under Independent Cascade Model in Online Social Networks

Abstract

This paper studies the problem of identifying rumor source in online social networks in which the spread of information follows the popular Independent Cascade model. In the absence of text information, we develop a monitor based approach to evaluate how likely that a piece of information is actually a rumor. Given the underlying social network structure, a number of monitor nodes are injected into the network whose job is to report the data they receive. Based on observing which of monitors received the information and which did not, we propose a polynomial time algorithm to compute rumor quantifier, a reachability based score for ranking the importance of nodes as the rumor source. Extensive simulation results have shown that, with a reasonable number of monitor nodes and appropriate monitor deployment, our rumor source detection algorithm can recognize rumor source effectively and efficiently.