Sources of misinformation in Online Social Networks: Who to suspect?

Abstract

Online Social Networks (OSNs) have recently emerged as one of the most effective channels for information sharing and discovery due to their ability of allowing users to read and create new content simultaneously. While this advantage provides users more rooms to decide which content to follow, it also makes OSNs fertile grounds for the wide spread of misinformation which can lead to undesirable consequences. In order to guarantee the trustworthiness of content sharing in OSNs, it is thus essential to have a strategic investigation on the first and foremost concern: the sources of misinformation. In this paper, we study k-Suspector problem which aims to identify the top k most suspected sources of misinformation. We propose two effective approaches namely ranking-based and optimization-based algorithms. We further extend our solutions to cope with the incompleteness of collected data as well as multiple attacks, which mostly occur in reality. Experimental results on real-world datasets show that our approaches achieve competitive detection ratios in a timely manner in comparison with available methods.