Interoperable pipelines for social cyber-security: assessing Twitter information operations during NATO Trident Juncture
- PDF / 1,728,387 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 76 Downloads / 151 Views
Interoperable pipelines for social cyber‑security: assessing Twitter information operations during NATO Trident Juncture 2018 Joshua Uyheng1 · Thomas Magelinski1 · Ramon Villa‑Cox1 · Christine Sowa1 · Kathleen M. Carley1
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract Social cyber-security is an emergent field defining a multidisciplinary and multimethodological approach to studying and preserving the free and open exchange of information online. This work contributes to burgeoning scholarship in this field by advocating the use of interoperable pipelines of computational tools. We demonstrate the utility of such a pipeline in a case study of Twitter information operations during the NATO Trident Juncture Exercises in 2018. By integratively utilizing tools from machine learning, natural language processing, and dynamic network analysis, we uncover significant bot activity aiming to discredit NATO targeted to key allied nations. We further show how to extend such analysis through drill-down procedures on individual influencers and influential subnetworks. We reflect on the value of interoperable pipelines for accumulating and triangulating insights that enable social cyber-security analysts to draw relevant insights across various scales of granularity. Keywords Social cyber-security · Information operations · Interoperability
1 Introduction Online social networks have allowed people to exchange information and express their views at an unprecedented scale. But while some scholars have argued that such platforms democratize public discourse, recent years have shown how adversarial actors may employ diverse strategies to manipulate public opinion toward disruptive social and political outcomes. Information operations are not a new phenomenon. * Joshua Uyheng [email protected] 1
Center for Computational Analysis of Social and Organizational Systems (CASOS), Institute for Software Research, Carnegie Mellon University, Wean Hall, 5000 Forbes Ave., Pittsburgh, PA 15213, USA
13
Vol.:(0123456789)
J. Uyheng et al.
However, their adaptation to online settings uniquely capable of influencing large collectives has made them a more urgent area of concern. Such activities have been documented across high-stakes contexts such as national elections, potentially even posing a threat to international security. How might researchers study information operations in a comprehensive and informative manner? Significant prior work has attempted to tackle this question from a variety of perspectives (Shu et al. 2017; Tucker et al. 2018; Zhou et al. 2019). While prevailing scholarship has primarily focused on developing state-of-the-art tools for solving social cyber-security problems one by one, scant work demonstrates how such tools may be integrated in practice. In the context of real-world online information operations, we argue that the utility of single tools is limited. In contrast, methodological pipelines emphasizing the interoperability of such tools may yield multifaceted insigh
Data Loading...
