Research Summary: Birds of a Feather Flock Together: Network-Based Detection of Coordinated Disinformation Campaigns on Telegram

Disinformation continues to proliferate across social networks, exacerbated by shifts in content moderation policies. Traditional methods for identifying disinformation sources struggle with the emergence of new accounts. A novel graph-based methodology has been developed to detect previously unknown Telegram accounts engaged in spreading disinformation, focusing on structural interconnections rather than language-specific content.

Why it matters

The persistent and evolving threat of disinformation undermines public trust, impacts decision-making, and can destabilize various sectors. Effective detection mechanisms are crucial for maintaining information integrity and mitigating the societal and operational risks posed by coordinated campaigns.

Key insights

• Disinformation is increasingly prevalent across social networks.
• Recent actions, such as the removal of fact-checkers and safety councils by major platforms, suggest an ongoing escalation of disinformation.
• Tracking disinformation sources is challenging due to the constant emergence of new accounts.
• A graph-based, language-agnostic methodology has been proposed to identify unknown Telegram accounts disseminating disinformation.
• The methodology initiates with known disinformation groups and analyzes their interconnections to discover new contributors to false narratives.

Source

arXiv — Computers and Society — https://arxiv.org/abs/2610.00222