Disinformation detection in next-generation social media

Identifying Disinformation on Fediverse Social Media Platforms

The Disinformation Detection in Next-Generation Social Media service includes two key modules.

The Fediverse Network Moderation Tool identifies potentially harmful servers in the Fediverse network by categorizing banned servers based on keywords and providing an updated list through an intuitive interface.

The Mastodon Credibility Service analyzes the credibility of information on Mastodon instances using data from message reshares, instance bans, and content analysis. It offers tools for instance inspection and visual reports on information dissemination trends.

This service connects with the Trustability and Credibility Service, as it analyzes social media networks or accounts, focusing on federated network specifics. While both services may share tools or modules (depending on technical specifications), the Disinformation Detection Service applies these tools at an aggregated level rather than on individual text.

Leveraging advanced AI, the service accelerates the detection of potential foreign influence operations on the Fediverse. By analyzing message resharing patterns, moderation actions, and flagged content, AI uncovers coordinated behaviors that may manipulate public discourse. It enables the automated detection of misinformation clusters, tracing information flows, and identifying key players in dissemination networks. Journalists and fact-checkers can then quickly pinpoint suspicious activities and investigate narrative origins, addressing influence campaigns more efficiently.