The coordination network toolkit: a framework for detecting and analysing coordinated behaviour on social media - Journal of Computational Social Science
link.springer.com
This paper introduces and evaluates the Coordination Network Toolkit, an open-source software package and methodological framework designed to detect and analyse coordinated behaviour on social media platforms. As the dynamics of online communication continue to evolve, coordination analysis has emerged as an important field of study with significant implications for understanding online influence, digital astroturfing, and online activism. Recognising the absence of a comprehensive, open-source tool for constructing coordination networks, our approach fills this gap, catering to multiple behaviors across diverse social media platforms. Our approach synthesises and significantly enhances various methods to provide a methodological framework for ‘multi-behaviour’ coordination detection, utilising weighted, directed multigraphs to capture intricate coordination dynamics. We evaluate our approach by revisiting a case study of the 2020 #ReopenAmerica Covid protest movement on Twitter. The paper concludes with a set of recommendations for future work, emphasising the need for a tailored statistical framework for coordination analysis and a deeper exploration into the motives behind online coordination.

The authors introduce and evaluate an open-source software package and methodological framework for detecting and analysing coordinated behaviour on social media, namely the Coordination Network Toolkit, utilising weighted, directed multigraphs to capture intricate coordination dynamics.

To whom it may concern.

    • 0x815@feddit.deOP
      0·
      6 months ago

      This is maybe a good idea. What would an emoji analysis tell us about a network? 😃

      • Jdreben@beehaw.org
        0·
        6 months ago

        There’s a South Park episode where they use “Emoji Analysis” to determine the identities of internet trolls. I am not original commenter but I laughed when I read their comment because indeed this research is a bit similar to the trend analysis they do in that episode.

  • MagicShel@programming.dev
    0·
    6 months ago

    I’ve always wanted to do this sort of thing, but lack the math and research background to do a good job of it. Glad someone is working on it.