FACTFLOW GUARDIAN: MINIMIZING ACTIVITY OF MISINFORMATION

Abstract

In recent times, the proliferation of online social media has transformed the information
landscape, altering the dynamics of information access for individuals. This surge,
however, has also ushered in challenges, particularly in the veracity of information, as
various forms of misinformation proliferate rapidly through these social platforms.
Ensuring effective governance of the online space and establishing a trusted network
environment have become imperatives. This article delves into anovel problem
termed the "Activity Minimization of Misinformation Influence" (AMMI) problem,
aiming to identify a set of nodes in a network that, when blocked, minimizes the total
amount of misinformation interaction between nodes (TAMIN).In essence, the AMMI
problem seeks to strategically select K nodes within a given social network G to
mitigate the overall TAMIN. We establish that the objective function lacks both
submodular and supermodular properties and introduce a heuristicgreedy algorithm
(HGA) designed to select the top K nodes for removal. To assess theeffectiveness of
our proposed method, comprehensive experiments have been conducted on three realworld
networks. The results affirm that our method surpasses comparison approaches,
providing a promising strategy for mitigating the impact of misinformation within
social networks.