Authors: Andrea Failla, Giulio Rossetti

Published on: April 29, 2024

Impact Score: 7.2

Arxiv code: Arxiv:2404.18984

Summary

• What is new: Provides a large, high-coverage dataset from Bluesky Social covering social interactions and user-generated content, including the complete post history of over 4M users and the output of content recommendation algorithms.
• Why this is important: The reduction in access to social media APIs is hindering the advancement of computational social science by creating a shortage of recent, publicly available social media data.
• What the research proposes: A large dataset from Bluesky Social is introduced, including comprehensive user-generated content and interaction data, to aid in computational social science research.
• Results: The dataset allows for unprecedented analysis of online behavior, human-machine engagement patterns, and provides ground-truth data for studying content exposure effects, self-selection, and content virality and diffusion.

Technical Details

Technological frameworks used: Bluesky Social data collection and analysis

Models used: Content recommendation algorithms analysis, user interaction patterns

Data used: 235M posts from 4M users, social interaction data, output of content recommendation algorithms

Potential Impact

Social media platforms, online content recommendation services, digital marketing agencies

Want to implement this idea in a business?

We have generated a startup concept here: ViralityGenie.