Uncovering Coordinated Networks on Social Media: Methods and Case Studies

Abstract

Coordinated campaigns are used to influence and manipulate social media platforms and their users, a critical challenge to the free exchange of information online. Here we introduce a general network-based framework to uncover groups of accounts that are likely coordinated. The proposed method construct coordination networks based on arbitrary behavioral traces shared among accounts. We present five case studies of influence campaigns in the diverse contexts of U.S. elections, Hong Kong protests, the Syrian civil war, and cryptocurrencies. In each of these cases, we detect networks of coordinated Twitter accounts by examining their identities, images, hashtag sequences, retweets, and temporal patterns. The proposed framework proves to be broadly applicable to uncover different kinds of coordination across information warfare scenarios.

Malicious and Low Credibility URLs on Twitter During the AstraZeneca COVID-19 Vaccine Development

“…As reported previously [12,16], there are signs of coordination to spread the disinformation content. In this work, one of our assumptions is that the coordination is based on the content being shared (i.e., URLs mentioned in the tweets).…”

Replication Data for: VaccinItaly: monitoring Italian conversations around vaccines on Twitter

“…In particular, we believe that researchers can exploit our data to further pursue several directions. To name a few: investigate the prevalence of reliable and unreliable on-line information and their impact on real-world initiatives (Yang et al 2020); assess the presence of "coordinated inauthentic behaviour" and understand the role played by online malicious actors in such a critical moment (Pacheco et al 2020;Broniatowski et al 2018); analyze the polarization which takes place between pro and anti views on vaccines and understand the mechanisms which lead to the formation of hyper-partisan on-line communities (Schmidt et al 2018;Johnson et al 2020).…”

On the dynamics of political discussions on Instagram: A network perspective

“…Cruickshank et al [57] analyzed interactions among topics on Twitter by modeling a sequence of networks from co-occurrences of hashtags used by Tweeter users. Aiming to reveal coordinated behavior on Twitter, Pacheco et al [58] proposed a set of network models that capture different patterns of co-interactions among users. Examples of patterns include using similar hashtags, sharing the same images or chronological use of the platform.…”

“…Specifically, we adopt an approach that reveals edges in the projected network that, in fact, unveil how the discussion takes place on Instagram. In contrast to prior work [56,58,61,59,60], we remove those co-interactions formed by chance, due to the frequent heavy tail nature of the content and user popularity in social media [25]. To address this challenge, we propose a generative model to filter such noisy edges out of the network, thus retaining only salient edges in the network backbone.…”

“…Some of the simplest methods of backbone extraction explore topological metrics and global properties of the graph. They typically propose to select the most important edges for the study [65,59,58,56,61]. Examples include the use of Kcore searches for dense subgraphs [66,67] and the removal of edges based on a global threshold τ, either applied directly to edge weights [68,69,70] or to more sophisticated metrics such as the neighborhood overlap of a pair of nodes [71,72].…”