Fedivertex: a Graph Dataset based on Decentralized Social Networks for Trustworthy Machine Learning
Marc Damie, Edwige Cyffers
TL;DR
Fedivertex delivers a first public, multi-platform graph dataset derived from the Fediverse, enabling reproducible benchmarks for decentralized machine learning and time-evolving graph analysis. By collecting 182 graphs across seven software platforms with 14 weekly snapshots, and providing a Python loading package, the work provides a credible, real-world benchmark beyond traditional for-profit social networks. It introduces defederation prediction as a new task, demonstrates application to decentralized learning scenarios, and compares Fedivertex graphs to existing datasets to highlight unique structural dynamics. The dataset supports tasks in privacy-aware learning, community detection, and edge deletion dynamics, with ethical crawling and privacy-conscious design. This resource aims to advance trustworthy decentralized ML by offering diverse, time-evolving graphs reflective of real-world federation dynamics.
Abstract
Decentralized machine learning - where each client keeps its own data locally and uses its own computational resources to collaboratively train a model by exchanging peer-to-peer messages - is increasingly popular, as it enables better scalability and control over the data. A major challenge in this setting is that learning dynamics depend on the topology of the communication graph, which motivates the use of real graph datasets for benchmarking decentralized algorithms. Unfortunately, existing graph datasets are largely limited to for-profit social networks crawled at a fixed point in time and often collected at the user scale, where links are heavily influenced by the platform and its recommendation algorithms. The Fediverse, which includes several free and open-source decentralized social media platforms such as Mastodon, Misskey, and Lemmy, offers an interesting real-world alternative. We introduce Fedivertex, a new dataset of 182 graphs, covering seven social networks from the Fediverse, crawled weekly over 14 weeks. We release the dataset along with a Python package to facilitate its use, and illustrate its utility on several tasks, including a new defederation task, which captures a process of link deletion observed on these networks.