GATV Researchers Publish New Study on Bot Detection in Decentralised Social Networks
Researchers from the Visual Telecommunications Applications Group (GATV) at the Universidad Politécnica de Madrid have contributed to the scientific publication “Bots into the Fediverse”, recently published in Springer Nature’s Social Network Analysis and Mining journal.
The study, developed within the framework of the European AI-CODE project, addresses one of the emerging challenges of decentralised social networks such as Mastodon and Bluesky: the detection of automated accounts, or bots, that may influence the spread of information and disinformation.
The research introduces an artificial intelligence-based model capable of identifying different types of accounts using profile metadata and content-related features. The results demonstrate the effectiveness of these approaches in detecting automated activity across federated platforms with high accuracy, contributing to the development of more reliable tools for improving trust and security in digital ecosystems.
The study was carried out by Francisco Moreno García, Pablo Perdomo-Quinteiro, Gustavo Hernández-Peñaloza, Federico Álvarez García, Alberto Belmonte, and Miguel Antonio Barbero-Álvarez.
This publication further strengthens GATV’s contribution to European research initiatives focused on artificial intelligence, social network analysis, and the fight against disinformation.
The full article is available at:
