Building a Trusted Network of Energy Experts on 2 Twitter, Through Graph Traversal Strategies and Active 3 Node Classification


Author(s): Vincenzo De Leo*, Michelangelo Puliga, Martina Erba, Cesare Scalia, Andrea Filetti and Alessandro Chessa

In this study we analyze the Twitter (now X) Friendship Network, focusing on users relevant to the energy sector, spotlighting experts, professionals, and businesses connected as ‘following’. By analyzing their connections, we identify clusters within this network, revealing how they are grouped based on their roles. We show how the natural formation of these clusters on social media platforms like Twitter, significantly impacts public discourse around energy and related critical issues such as climate change. We also highlight how the dynamical interplay of misinformation leads to the formation of polarized user groups that often result in disengagement from online discussions. These clusters define small groups with shared ways of communicating. Unlike broader networks, information exchanges here are sparse, typically involving accounts set up for precise business aims. Moreover, we exploit a Machine Learning approach to detect key members in these specialized groups and to uncover how thesegroups stay connected. This approach let us gain insights into corporate communication on social media, offering a fresh view on professional networking. Our findings highlight how companies in the energy sector use Twitter to coordinate their activities, with key institutions playing a central role in keeping these networks organized.